Interventions For Fatigue in Peripheral Neuropathy

Interventions for fatigue in peripheral neuropathy (Review)
White CM, van Doorn PA, Garssen MPJ, Stockley RC
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2014, Issue 12
http://www.thecochranelibrary.com

Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . . 4
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
ADDITIONAL SUMMARY OF FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . 15
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 36
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Interventions for fatigue in peripheral neuropathy (Review) i

[Intervention Review]
Interventions for fatigue in peripheral neuropathy
Claire M White1 , Pieter A van Doorn2 , Marcel PJ Garssen3 , Rachel C Stockley4
1 Health & Social Care Research, Faculty of Life Sciences & Medicine, King’s College London, London, UK. 2 Department of Neurology,
Erasmus University Medical Center, Rotterdam, Netherlands. 3 Neurology, Jeroen Bosch Hospital And Erasmus Medical Centre, s’-
Hertogenbosch, Netherlands. 4 Physiotherapy, Faculty of Health, Psychology & Social Care, Elizabeth Gaskell Campus, Manchester,
UK
Contact address: Claire M White, Health & Social Care Research, Faculty of Life Sciences & Medicine, King’s College London, Room
3.22, Shepherd’s House, Guy’s Campus, London, SE1 1UL, UK. Claire.white@kcl.ac.uk.
Editorial group: Cochrane Neuromuscular Disease Group.

Publication status and date: Edited (no change to conclusions), published in Issue 12, 2014.
Review content assessed as up-to-date: 5 November 2013.
Citation: White CM, van Doorn PA, Garssen MPJ, Stockley RC. Interventions for fatigue in peripheral neuropathy. Cochrane Database
of Systematic Reviews 2014, Issue 12. Art. No.: CD008146. DOI: 10.1002/14651858.CD008146.pub2.
ABSTRACT
Background
Persistent feelings of fatigue (or subjective fatigue), which may be experienced in the absence of physiological factors, affect many people
with peripheral neuropathy. A variety of interventions for subjective fatigue are available, but little is known about their efficacy or the
likelihood of any adverse effects for people with peripheral neuropathy.
Objectives
To assess the effects of drugs and physical, psychological or behavioural interventions for fatigue in adults or children with peripheral
neuropathy.
Search methods
On 5 November 2013, we searched the Cochrane Neuromuscular Disease Group Specialized Register, CENTRAL, MEDLINE,
EMBASE, CINAHL Plus, LILACS and AMED. We also searched reference lists of all studies identified for inclusion and relevant
reviews, and contacted the authors of included studies and known experts in the field to identify additional published or unpublished
data. We also searched trials registries for ongoing studies.
Selection criteria
We considered for inclusion randomised controlled trials (RCTs) and quasi-RCTs comparing any form of intervention for fatigue
management in adults with peripheral neuropathy with placebo, no intervention or an alternative form of intervention for fatigue.
Interventions considered included drugs, pacing and grading of physical activity, general or specific exercise, compensatory strategies
such as orthotics, relaxation, counselling, cognitive and educational strategies.
Data collection and analysis
Two review authors independently assessed risk of bias and extracted study data. We contacted study authors for additional information.
We collected information on adverse events from the included trials.
Interventions for fatigue in peripheral neuropathy (Review) 1
Main results
The review includes three trials, which were all at low risk of bias, involving 530 people with peripheral neuropathy. The effects
of amantadine from one randomised, double-blind, placebo-controlled, cross-over trial comparing amantadine with placebo for the
treatment of fatigue in 80 people with Guillain-Barré syndrome (GBS) were uncertain for the proportion of people achieving a favourable
outcome six weeks post-intervention (odds ratio (OR) 0.56 (95% confidence interval (CI) 0.22 to 1.35, N = 74, P = 0.16). We assessed
the quality of this evidence as low. Two parallel-group randomised double-blind, placebo-controlled trials comparing the effects of two
doses of ascorbic acid with placebo for reducing fatigue in adults with Charcot-Marie-Tooth disease type 1A (CMT1A) showed that the
effects of ascorbic acid at either dose are probably small (standardised mean difference (SMD) -0.12 (95% CI -0.32 to 0.08, n = 404,
P = 0.25)) for change in fatigue after 12 to 24 months (moderate quality evidence). Neither ascorbic acid study measured fatigue at
four to 12 weeks, which was our primary outcome measure. No serious adverse events were reported with amantadine. Serious adverse
events were reported in the trials of ascorbic acid. However,risk of serious adverse events was similar with ascorbic acid and placebo.
Authors’ conclusions
One small imprecise study in people with GBS showed uncertain effects of amantadine on fatigue. In two studies in people with
CMT1A there is moderate-quality evidence that ascorbic acid has little meaningful benefit on fatigue. Information about adverse effects
was limited, although both treatments appear to be well tolerated and safe in these conditions.
There was no evidence available from RCTs to evaluate the effect of other drugs or other interventions for fatigue in either GBS,
CMT1A or other causes of peripheral neuropathy. The cost effectiveness of different interventions should also be considered in future
randomised clinical trials.
PLAIN LANGUAGE SUMMARY

Treatments for fatigue in peripheral neuropathy
Review question
To assess the effects of treatments for fatigue in people with peripheral neuropathy.
Background
Peripheral neuropathy is damage to the nerves outside the brain and spinal cord. Many people with peripheral neuropathy have feelings
of severe tiredness (fatigue) that are not necessarily related to physical problems such as muscle weakness. Evidence in other long-term
illnesses where fatigue is a problem suggests that medicines and other forms of treatment may help. The aim of this review was to assess
the effect of drugs and other treatments, such as general or specific exercise, orthotics (devices such as braces), relaxation, counselling
and cognitive, behavioural and educational strategies on feelings of fatigue in people with peripheral neuropathy.
Study characteristics
From a wide search, we identified three RCTs that met our selection criteria. The trials involved a total of 530 adults with peripheral
neuropathy. Treatments were ascorbic acid (vitamin C) in people with Charcot-Marie-Tooth disease in two trials, and amantadine
in Guillain-Barré syndrome in the third. Charcot-Marie-Tooth disease is an inherited nerve disease and Guillain-Barré syndrome is
a condition where there is inflammation of the peripheral nerves. We chose measures of fatigue at four to 12 weeks as our preferred
measure of the effects of treatment. The amantadine trial but not the ascorbic acid trials provided data at this time point; the ascorbic
acid trials provided data more than 12 months after the start of the intervention.
Key results and quality of the evidence
There was insufficient evidence to determine the effects of amantadine when compared with an inactive treatment (placebo) for fatigue.
and there were no major unwanted effects. We found evidence that ascorbic acid probably has little meaningful benefit for fatigue.
No major unwanted effects were identified, but the trials were small. We assessed the quality of this evidence as moderate because the
results were imprecise, which means they do not rule out the possibility that the drugs could have an effect. We found no evidence
from RCTs on other medicines or treatments for fatigue in peripheral neuropathy.
There is insufficient evidence to determine the effect of amantadine for fatigue in GBS and ascorbic acid for fatigue in CMT1A. More
high-quality studies are needed to provide evidence on which to base management of feelings of fatigue in peripheral neuropathy. The
cost effectiveness of different interventions should also be considered in future RCTs.
The evidence in this review is current to November 2013.

Interventions for fatigue in peripheral neuropathy (Review) S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Amantadine versus placebo for fatigue in Guillain-Barré syndrome
Patient or population: people with fatigue in Guillain-Barré syndrome

Settings:
Intervention: amantadine versus placebo
Outcomes Illustrative comparative risks* (95% CI) Relative effect No of participants Quality of the evidence Comments
(95% CI) (studies) (GRADE)
Assumed risk Corresponding risk
Placebo Amantadine
Reduction in fatigue of 257 per 1000 162 per 1000 OR 0.56 74 ⊕⊕ -

at least 1-point on the (71 to 318) (0.22 to 1.35) (1 cross-over study) low1
Fatigue Severity Scale
between 4 to 12 weeks
after commencement of
intervention
(FSS) from 1 to 7 (higher
score is worse)
Follow-up: mean 12
weeks
Change in fatigue 12 or - - - - - -
more weeks after com-
mencement of interven-
tion (not measured)
Change in activity limi- - - - - - -

tation 12 or more weeks
intervention (not mea-
sured)
4
Change in participation The mean change in par- The mean change in par- 74 ⊕⊕ -
12 or more weeks after ticipation 12 or more ticipation 12 or more (1 cross-over study) low1
commencement of inter- weeks after commence- weeks after commence-
vention using the ment of intervention using ment of intervention us-
Rotterdam Hand- the control group was ing the intervention group
icap Scale. Scale from: 0 0.34 points was
to 36. 0.31 higher
Follow-up: mean 12 (0.09 lower to 0.72
weeks higher)
Health related quality of - - - - - EHQ and SF-36 measured

life after 12 or more but data not fully reported
weeks or more after
commencement of inter-
vention (not reported)
Serious adverse events See comment See comment Not estimable 80 See comment 0 participants had seri-
Follow-up: mean 12 (1 cross-over study) ous adverse events in this
weeks study
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the
assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio; OR: odds ratio
GRADE Working Group grades of evidence

High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
1 Evidence downgraded two levels due to imprecision (small sample size and wide CI that crosses the line of no effect).
5
BACKGROUND an avoidance of physical activity and physical deconditioning, thus
further exacerbating fatigue (Moss-Morris 2005). Environmental
factors may also precipitate fatigue, since many people with pe-
Description of the condition ripheral neuropathies have to utilise compensatory strategies such
as orthoses or walking aids as a result of deformities or sensory
Peripheral neuropathies collectively affect about 2.4% of the pop-
deficits. Therefore, the experience of fatigue may be clearly related
ulation and may be either genetic or acquired, and either acute
to known underlying factors or, more commonly, be only partly
or chronic in nature (Martyn 1998). The pathological process af-
explained by these. Finally, measurement of the level of fatigue is
fects peripheral nerves, resulting in neurological damage to either
difficult both because the subjective experience of fatigue varies
the axon (ie, degeneration of the central nerve fibre), the myelin
between individuals and because measures of different types of
(ie, demyelination or destruction of the insulating sheath of the
fatigue are numerous.
nerve), or a combination of both. Recovery may be by remyelina-
tion, through which improvements in function can be rapid, and
in some people there is almost complete recovery. However, re-
generation of damaged axons usually takes many months or years Description of the intervention
and recovery, if present, may be incomplete (Tamura 2007). The evidence about the efficacy of interventions for fatigue in
Common symptoms of peripheral neuropathy include numbness, peripheral neuropathy and other chronic conditions is limited and
diminished or altered sensation (pins and needles), muscle weak- unclear. Interventions aim to address physical, environmental and
ness, and autonomic dysfunction. People may also experience fa- psychological factors contributing to fatigue, and include drugs,
tigue, pain, psychological dysfunction and poor social adjustment pacing and grading of physical activity, general or specific exercise,
(Lennon 1993; Pfeiffer 2001). Even when neurological function orthotics, relaxation, counselling, cognitive behavioural strategies
improves, residual symptoms often persist and fatigue is frequently and others.
an ongoing problem (Merkies 1999).
The experience of fatigue (or subjective fatigue) has been described
as a “an overwhelming sense of tiredness, lack of energy and feeling
of exhaustion” that is “not relieved by rest” and is a common sequel How the intervention might work
of chronic conditions (Bleijenberg 2003; Karlsen 1999; Krupp The presumed mode of action of drug interventions such as aman-
2003). Physiological fatigue, or “the loss of voluntary force-pro- tadine for generalised fatigue is unknown (Pucci 2009); one ran-
ducing capacity during exercise” (Bigland-Ritchie 1978) can con- domised controlled trial (RCT) showed no effect in inflamma-
tribute to subjective fatigue, but people can experience subjective tory peripheral neuropathy (Garssen 2006). It is likely that factors
fatigue in the absence of physiological factors and physiological contributing to subjective fatigue in individuals with peripheral
fatigue does not necessarily result in subjective fatigue. neuropathy must be clearly understood in order to target interven-
tions effectively. For example, where fatigue is due to weakness,
Physical factors such as residual muscle weakness mean that peo- strengthening exercise may be beneficial, but where altered mood
ple with peripheral neuropathy have to work harder during ev- or well-being is a factor then drug, cognitive or behavioural inter-
eryday activities than healthy individuals, which will feel more ventions may be more relevant. It is likely that multidimensional
of an effort. Axonal degeneration prevents nerve impulses from interventions that address individuals’ self perception and health
being conducted along the peripheral nerve and demyelination beliefs combined with strategies to increase levels of physical activ-
can lead to conduction block that can be exacerbated by muscle ity or participation in regular exercise or both may also be helpful.
activity (Cappelen-Smith 2000), leading to fatigue (Kaji 2000).
Although there is no direct problem with the muscles themselves,
it has been shown that people with peripheral neuropathy may be
unable to voluntarily activate their muscles fully, possibly because Why it is important to do this review
of dysfunction within the central nervous system (Garssen 2007; Fatigue is a frequent and often severe problem that affects everyday
Schillings 2003). This form of physiological fatigue is known activities and quality of life for people with peripheral neuropa-
as central fatigue and could possibly be due to reduced drive thy. Although there have been some RCTs of drug and non-drug
from nerve cells in the brain (cortical neurons) or to the effect interventions for fatigue in peripheral neuropathy, we know of no
of alterations in the number of surviving peripheral nerve axons systematic review.
and an increase in the number of muscle fibres they innervate
(Martinez-Figuerora 1977; Sanders 1996). In addition, psycho-
logical factors such as lack of motivation or low mood can also af-
fect voluntary activation of muscles (Gandevia 2001; Kent-Braun
1993). Fear of precipitating fatigue may also contribute and lead to OBJECTIVES

To assess the effects of drugs and physical, psychological or be- studies used more than one scale, we used the preferred ranking of
havioural interventions for fatigue in adults or children with pe- commonly-used scales shown below to identify the scale to be used
ripheral neuropathy. in the primary outcome analysis. Instruments that we considered
appropriate included: Fatigue Severity Scale (FSS) (Krupp 1989),
Chalder Fatigue Scale (Chalder 1993), Fatigue Impact Scale or
METHODS daily-FIS (D-FIS) (Fisk 1994; Fisk 2002), Visual Analogue Scale
of Fatigue (VAS-F) (Lee 1991), and Piper Fatigue Scale (PFS)
(Piper 1998).
Criteria for considering studies for this review
Secondary outcomes
1. Fatigue after 12 or more weeks.
Types of studies
2. Activity limitations after 12 or more weeks.
We included all randomised controlled trials (RCTs) or quasi- 3. Participation restrictions after 12 or more weeks.
RCTs of any drug or non-drug intervention to treat fatigue as- 4. Health-related quality of life after 12 or more weeks.
sociated with peripheral neuropathy, compared with placebo, no 5. Adverse events: any adverse events, adverse events which lead
treatment or other drug or non-drug interventions for fatigue. to discontinuation of treatment and serious adverse events, which
Quasi-RCTs are those in which randomisation was intended but are those which are fatal, life-threatening, or require prolonged
they use methodology that may be biased, such as alternation, or hospitalisation.
use of case record numbers or date of birth.
Types of participants Search methods for identification of studies

We considered for inclusion in the review all trials in which partic-
ipants (adults or children) had fatigue associated with a diagnosis
of peripheral polyneuropathy, including sensory, motor and com- Electronic searches
bined sensory and motor neuropathies. We did not consider trials On 5 November 2013, we searched the Cochrane Neuromuscular
including mostly people with focal disease such as local entrap- Disease Group Specialized Register, the Cochrane Central Register
ment neuropathies with pain as the primary presenting feature (for of Controlled Trials (CENTRAL) (2013, Issue 10), MEDLINE
example, cervical radiculopathy, carpal tunnel syndrome, brachial (January 1966 to October 2013), EMBASE (January 1980 to Oc-
plexus neuritis, etc), and poliomyelitis. We accepted the diagnosis tober 2013), LILACS (January 1982 to October 2013), CINAHL
of peripheral neuropathy offered by the study authors, provided Plus (January 1937 to October 2013) and AMED (January 1985
that it stipulated the presence of clinical impairment characteristic to October 2013). The detailed search strategies are in the ap-
of peripheral neuropathy. We did not include diagnoses depen- pendices: Appendix 1 (MEDLINE), Appendix 2 (EMBASE), Ap-
dent on symptoms suggestive of neuropathy alone or neurophys- pendix 3 (CINAHL Plus), Appendix 4 (CENTRAL), Appendix 5
iological abnormalities in the absence of clinical signs. (AMED), Appendix 6 (LILACS) and Appendix 7 (the Cochrane
Neuromuscular Disease Group Specialized Register).
Types of interventions On 12 June 2014 we searched the Current Controlled
Trials register (www.controlled-trials.com), ClinicalTrials.gov (
We considered trials including any form of intervention for fatigue
www.clinicaltrials.gov) and the WHO International Clinical Trials
management, such as drugs, pacing and grading of physical activ-
Registry Platform (ICTRP) (www.who.int/ictrp/en/), using the
ity, general or specific exercise, compensatory strategies such as or-
search terms ’fatigue’ and ’peripheral neuropathy’.
thotics, relaxation, counselling, cognitive and educational strate-
gies and others, compared with either placebo, no intervention or
an alternative form of intervention for fatigue. Searching other resources
We reviewed the bibliographies of the randomised trials identi-
Types of outcome measures fied, contacted the trial authors and known experts in the field,
and approached pharmaceutical companies to identify additional
published or unpublished data.
Primary outcomes
Our primary outcome was fatigue severity and symptoms as mea-
sured by any validated scale evaluated at least four weeks and less
Data collection and analysis
than 12 weeks after commencement of the intervention. Where

Selection of studies outcome in different ways we analysed results using standardised
Two review authors (CMW, RCS) independently examined the ti- mean differences (SMDs) with a 95% CI.
tles and abstracts identified by the search. There were no language
restrictions. We obtained English language translations where ap- Dichotomous data
propriate. We retrieved all trials that were relevant and reviewed
For studies using similar outcomes with dichotomous data we
full-text publications. We examined retrieved publications against
calculated a pooled estimate of the risk ratio (RR), with a 95% CI.
the selection criteria and we resolved any disagreement by con-
sensus.We did not have to resort to arbitration by a third author
(MPG). If trials had had a heterogenous sample of disorders, we Unit of analysis issues
would have only included them if data from participants with Where there were multiple intervention groups within a trial,
peripheral neuropathy could be isolated or if at least 75% of the we made pairwise comparisons of similar interventions or active
participants had peripheral neuropathy. components versus no treatment, placebo, or another interven-
tion. We planned to analyse cross-over trials with continuous out-
Data extraction and management come measures using generalised inverse variance analysis (GIV)
(DerSimonian 1986), where the estimated difference in the mean
For each retrieved publication, two review authors (CW and
treatment effects and their standard errors were available for pool-
RCS) independently extracted the relevant data using standardised
ing with equivalent results from parallel-group studies. If this in-
forms. We resolved any disagreements by discussion. We gathered
formation had not been available then we would have pooled data
data on:
from the first phase only, where possible, with parallel design stud-
• eligibility criteria;
ies. For cross-over studies with dichotomous outcomes we would
• interventions;
have pooled results using odds ratios (ORs) and confidence inter-
• details of participants;
vals from paired analyses where available (Elbourne 2002). Only
• assignment to groups;
one cross-over trial was included for any comparison, so no deci-
• outcome measures;
sions about pooling data were necessary. We would have reported
• time at which outcomes are measured;
trials with unique designs or outcomes without a meta-analysis.
• funding;
We excluded studies that were not RCTs from the analyses but we
• declarations of interest;
comment on them in the Discussion.
• sample size; and
• statistical analysis.
Dealing with missing data
Assessment of risk of bias in included studies For all outcomes, where continuous data were presented as mean
change with 95% CIs, the standard deviations (SDs) were cal-
Two out of three review authors (CW, RS and MG) independently
culated for input into analyses and forest plots assuming a nor-
assessed all included studies for risk of bias. We graded the items
mal distribution due to a moderate sample size, using the equa-
according to the Cochrane Handbook for Systematic Reviews of In-
tion SD = SQRT of number of participants X (upper limit of CI
terventions (Higgins 2011a) and presented our assessments in a
to lower limit of CI) / 3.92. as recommended in Section 7.7.3.2
’Risk of bias’ table. We assessed trials for random sequence gen-
of the Cochrane Handbook for Systematic Reviews of Interventions
eration, allocation concealment, blinding (participants, personnel
(Higgins 2011b). Where overall disability sum score (ODSS) data
and outcome assessors), incomplete outcome data, selective out-
were presented as median and interquartile ranges, we did not in-
come reporting and other potential sources of bias.
put any data for analysis. Where participant dropout led to miss-
ing data we contacted trial authors if an intention-to-treat analysis
We then made a judgement on each of these criteria relating to
had not been performed. We also contacted authors where there
the risk of bias, of ’low risk of bias’, ’high risk of bias’ or ’unclear
were missing data for review outcomes, and additional data were
risk of bias’.
provided for the secondary outcome of participation in one study
(Garssen 2006).
Measures of treatment effect
Assessment of heterogeneity
Continuous data We undertook meta-analysis when studies investigated similar in-
For studies using the same outcome measures, we summarised terventions, used similar outcome measures and included groups
continuous data using mean differences (MDs), sometimes called of participants who were clinically homogeneous. Where studies
’weighted mean differences’, with a corresponding 95% confi- were heterogeneous for some outcomes we undertook a narrative
dence interval (CI). When different studies measured the same review.

If studies were similar in intervention, outcome measurement and Sensitivity analysis
participants, we assessed possible inconsistency across studies us- Insufficient data available. See Appendix 8 for methods planned
ing the I² statistic (Higgins 2003; Higgins 2011a). If studies had but not implemented.
been heterogeneous (Q statistic = 0.1 and I² statistic of 25% or This review has a protocol (White 2009). We listed any deviations
greater), the review authors would have considered conducting from the protocol in Differences between protocol and review.
subgroup analysis only. If we had considered the primary studies
to be heterogeneous even within subgroups, we would have un-
dertaken a narrative approach rather than a meta-analysis.
RESULTS
Assessment of reporting biases
We did not use funnel plots to investigate associations between
effect size and study precision in terms of sample size, because Description of studies
of the small number of studies included in the review. If we had See Characteristics of included studies and Characteristics of
included funnel plots, we would have explored the relationships excluded studies for details.
observed to investigate differences between studies with large and The electronic and manual searches identified a total of 1386 ti-
small samples, or systematic biases, such as publication bias. tles and abstracts. After removing duplicates and excluding ab-
stracts where studies were clearly not eligible for inclusion, only
Data synthesis six full publications, four protocols and two abstracts were rele-
For studies with a similar type of intervention or a similar active vant to the review. After reading the full text of the six published
component, we performed a meta-analysis to calculate a treatment articles we excluded three: two were not randomised controlled
effect across trials with a fixed-effect model, as heterogeneity was trials (RCTs) (Carter 2006; Enderlin 2008) and one was an RCT
not demonstrated. Dichotomous outcome results were available of ascorbic acid for children with CMT1A in which fatigue was
from only one study so we did not present pooled RRs. Where it not an outcome (Burns 2009). We included one abstract of con-
was not possible to perform a meta-analysis we provided descrip- ference proceedings, regarding an RCT of a new low-dose com-
tive summaries of the results from each trial. bination of three already approved drugs for CMT1A, as an on-
going trial, as the analyses are ongoing (Attarian 2013). We will
include this study in future updates of this review if appropriate.
’Summary of findings’ table Two of the other five ongoing or unpublished studies were RCTs
We created ’Summary of findings’ tables using the following out- of ascorbic acid in CMT1, and we contacted the authors for fur-
comes: ther information (NCT00484510; Verhamme 2009). However,
1. Fatigue severity and symptoms between four and 12 weeks in these ongoing studies, fatigue was not included as an outcome
after commencement of the intervention. and we excluded them. One randomised trial of coenzyme Q10
2. Fatigue after 12 or more weeks. for CMT (NCT00541164), with changes in weakness, fatigue
3. Activity limitations after 12 or more weeks. and pain as the primary outcome, is completed but appears to be
4. Participation restrictions after 12 or more weeks. unpublished. NCT02121678 is not yet recruiting, but will study
5. Health-related quality of life after 12 or more weeks. resistance training for chronic inflammatory demyelinating neu-
6. Serious adverse events. ropathy (CIDP) with the fatigue severity scale (FSS) as a planned
We used the five GRADE considerations (study limitations, secondary outcome. We identified one study, Ramdharry 2011,
consistency of effect, imprecision, indirectness and publication from an abstract in the Cochrane Neuromuscular Disease Group
bias) to assess the quality of a body of evidence (studies that Specialized Register. The study appears to be ongoing and the au-
contribute data) for the prespecified outcomes. We used meth- thors provided no data when contacted. We will include it if eli-
ods and recommendations described in Chapters 11 and 12 of gible once the full publication is available. See Characteristics of
the Cochrane Handbook for Systematic Reviews of Interventions ongoing studies.
(Schünemann 2011a; Schünemann 2011b). We used GRADEpro The study selection process therefore resulted in only three trials
software (GRADEpro 2008). We explained decisions to down- or finally meeting the criteria for inclusion (Garssen 2006; Micallef
up-grade the quality of studies using footnotes and we made com- 2009; Pareyson 2011).
ments to aid reader’s understanding of the review where necessary.
Study design and participants
Subgroup analysis and investigation of heterogeneity We include three studies evaluating the efficacy of two drugs for
Insufficient data available. See Appendix 8 for methods planned fatigue in two different populations (Garssen 2006; Micallef 2009;
but not implemented. Pareyson 2011).

The first study (Garssen 2006) was a randomised, double-blind, into groups of four stratified by disease severity and centre was
placebo-controlled, cross-over trial of amantadine for severe fa- followed by 24 months of treatment with either ascorbic acid or
tigue in 80 neurologically well-recovered participants who had had placebo, and outcomes were assessed at six-monthly intervals from
Guillain-Barré syndrome (GBS). Randomisation was in blocks of commencement of intervention until 24 months. Six participants
six with half the study participants receiving amantadine first and did not receive the allocated intervention after randomisation, and
the remainder receiving placebo first. Interventions were followed are dropped from the denominators. A total of 22 participants
by a two-week washout period. Assessment of outcomes occurred withdrew from the intervention after commencement. Reasons for
on six occasions: 1: at baseline, 2: following two weeks of no in- withdrawal were adverse events (13), withdrawal of consent (six),
tervention, 3: after six weeks of intervention 1, 4: after a two-week moved away from study area (two) or for personal reasons (one).
washout period, 5: after six weeks of intervention 2, and finally Whilst the primary study outcome was based on all randomised
6: after a further two-week washout period. Three randomised patients who received at least one dose of the study drug using im-
participants were withdrawn from the study in the pre-cross-over putation for missing data the secondary analyses were conducted
phase. One participant was in the placebo group and withdrew be- without imputation of missing data.
cause of concern over potential side effects of amantadine when she
became pregnant. The two other participants were in the amanta-
Interventions
dine treatment group; one withdrew because of hospital admission
with acute cholangitis and the second participant developed severe The study of amantadine in Guillain-Barré syndrome was a cross-
complaints of dizziness. A further three participants were excluded over trial in which amantadine was compared with placebo at a
from the analysis due to incorrect completion of the FSS. There- dose of one tablet (100 mg) daily for one week followed by two
fore, 74 participants with GBS were included in the reporting of tablets daily for the remaining five weeks of each intervention
outcomes. All participants had been diagnosed with GBS between period. A washout period of two weeks was incorporated before
six months and 15 years previously and had no apparent changes the participants shifted to the other treatment arm (amantadine
in GBS disability score for the three months prior to entering the or placebo) (Garssen 2006).
study. All could walk at least 10 metres with or without a walking The studies of ascorbic acid in Charcot-Marie-Tooth disease 1A
aid (GBS disability score < 3). were a parallel-group study comparing 1 g or 3 g of ascorbic acid
The two remaining studies (Micallef 2009; Pareyson 2011) were in a single administration of three capsules with three placebo
parallel-group, randomised, double-blind, placebo-controlled tri- capsules for 12 months (Micallef 2009) and a parallel-group study
als of the effect of ascorbic acid in a total of 450 adults with Char- comparing 1.5 g of ascorbic acid in two daily doses with placebo
cot-Marie-Tooth disease type 1A (CMT1A). In one, two doses tablets for 24 months (Pareyson 2011).
(either 1 g or 3 g daily) of ascorbic acid or placebo were used in
the treatment of 179 participants (Micallef 2009). Confirmation Outcomes
of diagnosis was by clinical examination and genotyping with du-
In participants with severe fatigue due to Guillain-Barré syndrome
plication in 17p11.2 and at least one motor sign or symptom (gait
(defined as Fatigue Severity Scale (FSS) 5.0 or more; FSS range 1
disorder, distal amyotrophy, foot deformation or distal weakness).
to 7), Garssen 2006 used a reduction of at least one point on the
Block randomisation into three groups with matching for study
FSS as a primary outcome measure of fatigue. Outcomes for FSS
site and sex was followed by 12 months of treatment with either
were reported for both phases of the cross-over study. Secondary
1 g or 3 g of ascorbic acid or placebo; outcomes were assessed at
outcome measures in the major domains of the International Clas-
baseline and after 12 months of intervention. A total of 16 partici-
sification of Functioning, Disability and Health (ICF) were: for
pants withdrew from the study after randomisation due to adverse
structure and function, anxiety and depression using the Hospi-
events (six), withdrawal of consent (six), loss to follow-up (two),
tal Anxiety and Depression Scale (HADS); for activity, impact of
non-compliance (one) or missing baseline data (one).
fatigue using the Fatigue Impact Scale (FIS); and for participa-
Therefore, 163 participants completed the study but analysis was
tion, Rotterdam Handicap Scale (RHS), quality of life using the
of all 179 participants, on an intention-to-treat basis.
Medical Outcomes Short Form-36 (SF-36) Health Survey and the
In the second multicentre study (Pareyson 2011), the efficacy of
Euroquol Health Questionnaire (EQHQ); and adverse events.
chronic administration, for 24 months, of either 1.5 g/day ascor-
In the studies of ascorbic acid in participants with Charcot-Marie-
bic acid or placebo in the treatment of 271 participants (eligible
Tooth disease 1A, the primary outcome was CMT neuropathy
for analysis) with Charcot-Marie-Tooth disease 1A was evaluated.
score (CMTNS). Fatigue was assessed as a secondary outcome
Participants had a clinical diagnosis of Charcot-Marie-Tooth dis-
using the change in Visual Analogue Scale - Fatigue (VAS-F),
ease 1A with genetic confirmation of 17p11.2-p12 duplication
either 12 months (Micallef 2009) or 24 months (Pareyson 2011)
and a CMT neuropathy score of between 1 (excluding the elec-
after commencement of the intervention. Secondary outcomes of
trophysiological component) and 35 (including the electrophys-
activity limitation using the Overall Disability Sum Score (ODSS)
iological component). Block randomisation of 277 participants
(Micallef 2009) or Overall Neuropathy Limitations Scale (ONLS)

(Pareyson 2011), 10 metre walk test (10MWT) Pareyson 2011
and 9 hole peg test (9HPT) (Pareyson 2011) were reported, as
well as measures of participation and quality of life using the SF-
36 and adverse events in both studies.
Risk of bias in included studies

See Characteristics of included studies for details and Figure 1.
Figure 1. Methodological quality summary: review authors’ judgements about each methodological quality
item for each included study. Key: green (+) = low risk of bias; yellow (?) = unclear risk of bias; red (-) = high risk
of bias.

Garssen 2006 conducted block randomisation in blocks of six,
although allocation was concealed from researchers; we therefore this study to be at low risk of bias.
considered the study to be at low risk of bias for these criteria. The Effects of interventions
trial was a double-blind, placebo-controlled study, with 80 partici-
pants initially enrolled. Although the final analysis was based on 74 See: Summary of findings for the main comparison Amantadine
participants as six were withdrawn, there was adequate reporting versus placebo for fatigue in Guillain-Barré syndrome; Summary
of the loss to follow-up. Sample size was calculated using estimates of findings 2 Ascorbic acid versus placebo for fatigue in peripheral
of proportion from unpublished data obtained from a pilot study, neuropathy
indicating that 36 participants per group were required for 90%
power at a two-sided 5% alpha, so that with 77 participants com- Amantadine versus placebo in Guillain-Barré
pleting the trial it was sufficiently powered to detect a difference. syndrome
The washout period of two weeks between cross-over intervention Effect estimates, where data were available, were obtained from
phases seems sufficient to avoid a carry-over effect of amantadine, a single trial, so we downgraded the quality of evidence for all
and is comparable with other randomised controlled cross-over outcomes for this reason (Summary of findings for the main
designs of efficacy for amantadine in other populations (Cohen comparison).
1989; Pucci 2009; Rosenberg 1988). The authors reported data for
mean (SD) improvement in FSS scores for both amantadine and Primary outcome
placebo treatment in both the amantadine-placebo (0.46 (1.29), P
= 0.078; 95% CI -0.92 to 0.05) and placebo-amantadine sequence Fatigue between four and 12 weeks after commencement of
(0.08 (1.44), P = 0.28; 95% CI -1.24 to 0.44) groups and t-test intervention
analysis for period effects and carry-over effects were not signifi-
Garssen 2006 assessed baseline measures at visit 1 and at visit 2,
cant. There was some evidence of selective reporting of secondary
prior to randomisation. In the 74 participants for whom full out-
outcomes, as only P values were reported. Overall we considered
come data were reported, there was evidence of a reduction in fa-
the study to be at low risk of bias.
tigue between these baseline visits, which was stated by the authors
In Micallef 2009, sequence generation was conducted by an inde-
to approach significance (P = 0.05 Wilcoxon signed rank test), al-
pendent contract research organisation who were also responsible
though complete group data were not presented. The magnitude
for concealed allocation of randomisation into blocks of 12 that
of fatigue reduction before the start of any intervention was such
were stratified for sex and study site (three sites). The trial was dou-
that although the median FSS score for the 74 participants was 5.9
ble-blind with the administration of visually identical hard gelatin
at visit 2, seven (9%) of the participants had an FSS score lower
capsules for all groups. Analysis was on an intention-to-treat basis
than the inclusion criterion of FSS 5 for severe fatigue.
with 179 participants randomised and analysed despite 15 with-
Thirty-six participants received amantadine followed by placebo
drawals prior to follow-up (four from the placebo group, eight
and 38 received placebo followed by amantadine. Favourable out-
from the 1 g ascorbic acid group and three from the 3 g ascorbic
come reporting was available for both phases of each treatment se-
acid group); one participant from the placebo group had missing
quence, representing evaluation of the primary outcome six weeks
baseline data for the study’s primary outcome. All study outcomes
after the commencement of intervention. Results were presented
were adequately reported. Whilst authors stated that the data were
as the mean difference (MD) in reduction in FSS scores between
analysed on an intention-to-treat basis there was no indication of
amantadine and placebo, and as an odds ratio (OR) of a favourable
how the investigators dealt with the missing data from 16 partici-
outcome (defined as a reduction of more than 1 in FSS score). The
pants. Our overall summary assessment was that the study was at
overall MD in FSS score between amantadine and placebo phases
low risk of bias.
was -0.45 (95% CI -0.94 to 0.04; t = -1.80; df = 73; P = 0.076).
In the third study (Pareyson 2011), an independent randomisa-
When the participants responding favourably to one treatment
tion unit was responsible for sequence generation and group allo-
only in both the group receiving amantadine followed by placebo
cation stratified for disease severity and treatment centre in blocks
(amantadine-placebo group) and those receiving placebo followed
of four. This, combined with the use of placebo tablets of identical
by amantadine (placebo-amantadine group) were combined, the
appearance, taste and smell, resulted in blinding of participants,
number with a more favourable outcome following amantadine
researchers and treating physicians. Six participants randomised to
was 9 of 25 (36%) participants with an OR of 0.56 (95% CI 0.22
the intervention (four participants) and placebo (two participants)
to 1.35, P = 0.16) using the McNemar test.
groups withdrew consent before receiving the allocated interven-
tion and were excluded from analysis. Although the investigators
recorded outcome data at six-monthly intervals for 24 months, Secondary outcomes
the authors stated that their main analysis would be at 24 months,
All stated secondary outcomes in Garssen 2006 were only reported
and reported the primary outcome at only 12 and 24 months and
at six weeks after commencement of the intervention and therefore
the secondary outcomes only at 24 months. Overall we considered
not at the stated secondary outcome point of 12 or more weeks

after commencement of intervention for this review. In addition dine group and one in placebo group) and other less frequent side
full data for the study secondary outcomes were not available, as effects (headache, feelings of nervousness and vivid dreams). No
only P values were provided (Garssen 2006). The data indicated side effects were reported to have affected trial participation.
that there was no significant improvement in: fatigue impact us-
ing the fatigue impact scale (FIS) (P = 0.77); participation using
the Rotterdam Handicap Scale (RHS) (P = 0.14) and Euroquol
Health Questionnaire (EQHQ) (P = 0.21) at six weeks. The au- Ascorbic acid versus placebo in Charcot-Marie-Tooth
thors reported that quality of life using the SF-36 demonstrated a disease
significant improvement in physical role functioning (P = 0.008)
and mental health perception (P = 0.008), although only in the Primary outcome
placebo group (Garssen 2006). We contacted the authors for fur-
ther information about secondary outcomes; the RHS scores for Fatigue between four and 12 weeks after commencement of
participation at six weeks from the original data were provided intervention
by the trial author (Garssen 2006). The data were analysed us- Neither study evaluating ascorbic acid for Charcot-Marie-Tooth
ing a paired two-sample t-test and showed that the MD between disease 1A reported change in fatigue within the time scale of the
the change in participation with placebo versus the change with primary outcome for this review (Micallef 2009; Pareyson 2011).
amantadine between four to 12 weeks after commencement of
intervention was 0.31 (95% CI -0.09 to 0.72, P = 0.13) (Table 1).
(Summary of findings for the main comparison). Secondary outcomes
Fatigue after 12 or more weeks Data for secondary outcomes were reported at 12 months after
Not reported at this time point. commencement of the intervention in Micallef 2009 and at 24
Activity limitations after 12 or more weeks months after commencement of the intervention in Pareyson
Not reported at this time point. 2011. Where two different doses of ascorbic acid were used in
Participation restrictions after 12 or more weeks Micallef 2009, we combined data from both intervention arms,
Not reported at this time point. including participants receiving a higher dose (3 g ascorbic acid)
Health-related quality of life (HRQoL) after 12 or more weeks and those receiving a lower dose (1 g ascorbic acid) in comparisons
Not reported at this time point. for all outcomes.
Fatigue after more than 12 weeks

Adverse events
Reported in Micallef 2009 and Pareyson 2011.
Reported in Garssen 2006. Both studies assessed fatigue using change in Visual Analogue Scale
No serious adverse events were reported. However, mild, transient for fatigue (VAS-F). Fatigue was assessed at 12-month follow-up
adverse effects were reported in 32 out of 80 (40%) participants: 22 using a 0 to 100 VAS for fatigue in Micallef 2009 and at 24-month
participants during amantadine treatment and 10 during placebo follow-up using a 0 to 10 VAS for fatigue in Pareyson 2011. The
treatment. Adverse effects included anticholinergic complaints (six pooled estimate of effect of the change in fatigue more than 12
participants: three in each treatment group), gastrointestinal com- weeks after commencement of the intervention was standardised
plaints (10 participants: seven in placebo group and three in aman- mean difference (SMD) -0.12 (95% CI -0.32 to 0.08, P = 0.25;
tadine group), sleep complaints (10 participants: nine in amanta- N = 404; I² = 0%) (see Analysis 1.1; Figure 2).
Figure 2. Forest plot of comparison: 2 Ascorbic acid versus placebo, outcome: 2.1 Change in fatigue > 12
weeks after commencement of intervention at 12 months [VAS 0 - 100).

Activity limitations after 12 or more weeks 2009, and no significant difference (P = 0.99) in median change
in activity limitation between placebo and 1.5 g dose of ascorbic
Reported in Micallef 2009 and Pareyson 2011. acid in Pareyson 2011.
Activity limitation was assessed using the ODSS at 12 months after In addition, further measures of activity limitation were included
commencement of intervention in Micallef 2009 and the Over- in both studies. The 10MWT was measured in both Micallef 2009
all Neuropathy Limitations Scale (ONLS) 24 months after com- and Pareyson 2011 (at 12 months and 24 months, respectively)
mencement of intervention in Pareyson 2011. Pooling of data in and the pooled estimate of effect for change in time taken to
a meta-analysis was not possible since data for ordinal scale scores walk 10 metres more than 12 weeks after commencement of the
were presented as medians plus interquartile ranges (IQRs), and intervention was MD -0.39 (95% CI -1.04 to 0.26, P = 0.24; N =
there was no indication of whether the data were skewed. How- 406; I² = 0%) (Analysis 1.2; Figure 3). In addition the 9HPT was
ever, the authors reported that there was no significant difference measured in Pareyson 2011 at 24 months and showed that there
(P = 0.22) in median change in activity limitation between placebo was no significant effect of ascorbic acid (MD -0.74, 95% CI -
and 1 g or 3 g doses of ascorbic acid at 12 months in Micallef 1.60 to 0.12, P = 0.09; N = 224) (Analysis 1.3).
Figure 3. Forest plot of comparison: 1 Ascorbic acid versus placebo, outcome: 1.2 Change in 10 m timed
walk > 12 weeks after commencement of the intervention (measured at 12 to 24 months) [seconds].
Participation restrictions after 12 or more weeks

Not reported in Micallef 2009 or Pareyson 2011.
Health-related quality of life (HRQoL) after 12 or more

weeks
Reported as an outcome and assessed using the SF-36 at 12 months
after commencement of the intervention in Micallef 2009 and
24 months after commencement of the intervention in Pareyson
2011. We calculated a pooled estimate of the change in physical
functioning, as measured by the Physical Component Score (PCS)
in Micallef 2009 and the Physical Function Scale (PFS) of the SF-
36 in Pareyson 2011. Using data from 400 participants, the SMD
was 0.08 (95% CI -0.12 to 0.28) favouring placebo (see Figure
4, Analysis 1.4). There was no significant statistical heterogeneity,
despite the difference in duration of follow-up (I² = 0%, P = 0.67).
Figure 4. Forest plot of comparison: 2 Ascorbic acid versus placebo, outcome: 2.4 Change in physical
function score of SF-36.

Micallef 2009 reported HRQoL assessed by the Mental Compo-
nent Score of the SF-36 and for this too there was no significant depression, colitis and insomnia) as reasons for withdrawal from
difference (P = 0.77) between either dose of ascorbic acid versus the study.
placebo or in a combined analysis, between both doses of ascorbic In Pareyson 2011, 21 serious adverse events were documented
acid versus placebo, 12 months after the start of treatment (Anal- in 20 participants (eight events in seven of 138 participants in
ysis 1.5). the ascorbic acid group and 13 events in 13 of 133 participants
in the placebo group). Researchers considered all but one event,
Adverse events
where a woman in the placebo group was admitted to hospital
Reported in Micallef 2009 and Pareyson 2011. due to anaemia, as unrelated to the intervention. Thirteen partic-
In Micallef 2009, 896 adverse events were documented in 158 ipants experiencing adverse events were lost to follow-up (six in
(89%) participants (55 of 62 participants in the placebo group the placebo group and seven in the ascorbic acid group), and eight
(89%); 47 of 56 participants in the 1 g ascorbic acid group (89%), participants (six in the placebo group and two in the ascorbic acid
and 56 of 61 participants in the 3 g ascorbic acid group (92%)), of group) ceased treatment as a result of a serious adverse event. Five
which only 24 were serious adverse events. Serious adverse events of these eight continued with follow-up and three (two receiv-
were recorded in 22 participants (eight in the placebo group, seven ing placebo and one receiving ascorbic acid) withdrew from the
in the 1 g ascorbic acid group and seven in the 3 g ascorbic acid study. Adverse events in those who discontinued the interventions
group). Researchers deemed the majority of events (70%) not to included gastralgia, abdominal cramps or diarrhoea (9), surgery
be related to treatment and 208 (25%) to have a possible relation (4), renal colic (1), pregnancy (3), rectal bleeding (1), saccharine
to treatment, of which only 12 (11 listed in the report) were se- intolerance (1) and gum disease (1). The pooled risk ratio (RR)
vere: upper abdominal pain (2), dyspepsia (2), headache (1), in- for serious adverse events in studies of ascorbic acid for CMT1A
somnia (1), muscle spasm (1), musculoskeletal stiffness (1), rash was 0.70 (95% CI 0.39 to 1.27, P = 0.24; N = 450; I² = 0%)
(1), rhinitis (1) and sleep disturbance (1). Six participants (one in (see Analysis 1.6, Figure 5), indicating no significant difference in
the placebo group, four in the 1 g ascorbic acid group and one in serious adverse events between ascorbic acid and placebo.
the 3 g ascorbic acid group) gave adverse events (cystitis, myalgia,
Figure 5. Forest plot of comparison: 2 Ascorbic acid versus placebo, outcome: 2.6 Adverse events.

Interventions for fatigue in peripheral neuropathy (Review) A D D I T I O N A L S U M M A R Y O F F I N D I N G S [Explanation]
Ascorbic acid versus placebo for fatigue in peripheral neuropathy
Patient or population: people with fatigue in peripheral neuropathy

Settings:
Intervention: ascorbic acid versus placebo
Outcomes Illustrative comparative risks* (95% CI) Relative effect No of participants Quality of the evidence Comments
(95% CI) (studies) (GRADE)
Assumed risk Corresponding risk
Placebo Ascorbic acid
Reduction in fatigue of - - - - - -
at least 1 point on the
(FSS) between 4 to 12
weeks after commence-
ment of intervention (not
measured)
Change in fatigue 12 or - The mean change in fa- - 404 ⊕⊕⊕ -

more weeks after com- tigue 12 or more weeks (2 studies) moderate1
mencement of interven- after commencement of
tion intervention in the inter-
Visual Analogue Scale vention groups was
(VAS) of fatigue 0.12 standard deviations
Follow-up: mean 12-24 lower
months (0.32 lower to 0.08
higher)
Change in activity limita- See comment See comment Not estimable 406 - Data available were not
tion (ODSS/ONLS) 12 or (2 studies) suitable for meta-analysis
more weeks after com-
mencement of interven-
tion
16
Overall Neuropathy Lim-

itations Scale or Over-
all Disability Sum Score.
Scale from: 0 to 12.
Follow-up: 12-24 months
Change in 10 metre The mean changes in 10 The mean change in 10 406 ⊕⊕⊕ -
timed walk 12 or more metre timed walk 12 or metre timed walk 12 or (2 studies) moderate1
weeks after commence- more weeks after com- more weeks after com-
ment of intervention mencement of interven- mencement of interven-
Time taken to walk 10 tion in the control groups tion in the intervention
metres in seconds werea decrease of 0.40 groups was a decrease
Follow-up: 12-24 months seconds and an increase of
of 0.76 seconds2 0.39 seconds (1.04
faster to 0.26 slower)
Change in speed of com- The mean change in The mean change in Not estimable 224 ⊕⊕⊕ -
pletion of 9-hole peg test speed of completion of speed of completion of (1 study) moderate1
(in seconds) 12 or more the 9-hole peg test was the 9-hole peg test was
weeks after commence- an increase of 0.85 sec- a decrease of 0.74 sec-
ment of intervention onds in the control group onds
Time (in seconds) taken (1.60 faster to 0.12
to complete 9-hole peg slower)
test
Follow-up: mean 24
months
Health-related quality of - The mean health-related - 400 ⊕⊕⊕ -

life 12 or more weeks quality of life 12 or more (2 studies) moderate1
after commencement of weeks after commence-
the intervention ment of the intervention
Change in SF-36 physi- (change in SF-36 physi-
cal function score cal function score) in the
SF-36 Physical Function- intervention groups was
ing Scale and Physical 0.08 standard deviations
Component Score higher
Follow-up: 12-24 months (0.12 lower to 0.28
higher)
17
Change in participation - - - - - -
12 or more weeks after
commencement of inter-
vention (not measured)
Health-related quality of The mean change in The mean health-related - 123 ⊕⊕⊕ -
life 12 or more weeks health-related quality of quality of life 12 or more (1 study) moderate1
after commencement of life 12 or more weeks weeks after commence-
the intervention Change after commencement of ment of the intervention
in mental function score the intervention (change (change in mental func-
of SF-36 in mental function score tion score of SF-36) in the
SF-36 Mental Component of SF-36) in the control intervention groups was
Score group was 1.2 points 0.9 higher
Follow-up: mean 12 (2.65 lower to 4.45
months higher)
Serious adverse events 108 per 1000 76 per 1000 RR 0.70 450 ⊕⊕⊕⊕ -
Follow-up: 12 months (42 to 137) (0.39 to 1.27) (2 studies) moderate1
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the
assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence

High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
1 Only one or two studies included.
2 Assumed control group risks are the control group values in the two trials.
18
Only three RCTs (Garssen 2006; Micallef 2009; Pareyson 2011)
DISCUSSION
were eligible for inclusion in the review. One was of amantadine
Summary of main results for severe fatigue in Guillain-Barré syndrome and the others were
of three different doses of ascorbic acid for people with Charcot-
Our search identified three randomised controlled trials (RCTs)
Marie-Tooth disease 1A. RCTs evaluating the effect of other in-
that met the inclusion criteria. The trials evaluated the effect of two
terventions for fatigue, such as other drugs; pacing and grading of
different drugs on fatigue in two different populations of people
physical activity; general or specific exercise; compensatory strate-
with peripheral neuropathy. Two studies examined the effect of
gies such as orthotics; relaxation; counselling; cognitive and edu-
different doses of ascorbic acid in Charcot-Marie-Tooth disease
cational strategies; and interventions for fatigue that form part of a
1A, and therefore no combined analysis was possible.
multidimensional programme, or trials of interventions for other
The first study (Garssen 2006) was a block-randomised, double-
causes of peripheral neuropathy, were not available. None of the
blind, placebo-controlled, cross-over study of amantadine for se-
included studies considered the cost effectiveness of interventions
vere fatigue in 74 neurologically stable participants at least six
for fatigue in peripheral neuropathy
months after Guillain-Barré syndrome. We judged the risk of bias
of the trial to be low. The study reported a moderate reduction
in fatigue severity scale (FSS) score for participants treated with
amantadine but the confidence interval was wide and study size Quality of the evidence
relatively small, suggesting that the study may have been under- The overall risk of bias in the included trials (Garssen 2006;
powered to detect a difference; we therefore graded the quality of Micallef 2009; Pareyson 2011) was low.
the evidence as low. These data indicate that larger studies might Garssen 2006 reported a double-blind, placebo-controlled study,
improve the power to detect a reduction in risk of fatigue with where, despite a block randomisation procedure affecting bias in
amantadine. Nevertheless, the mean difference (MD) did not ex- sequence generation and incomplete reporting of secondary out-
ceed the minimal clinically important difference for change in comes, there was effective allocation concealment and P values for
FSS scores estimated for people with systemic lupus erythemato- secondary outcomes were available. Reporting of data for fatigue
sus (MD 0.6; 95% CI 0.3 to 0.9), indicating that it is likely that during both phases of the trial indicates that there was no signifi-
there is no effect of amantadine compared with placebo in the cant carry-over effect. We therefore judged the overall risk of bias
treatment of severe fatigue (see Summary of findings for the main to be low.
comparison). A significant decrease in fatigue was noted between Micallef 2009 reported a block-randomised, double-blind,
two pre-treatment baseline visits two weeks apart. Since the par- placebo-controlled study, where all assessed aspects of risk of bias
ticipants were considered to be neurologically stable, this suggests were adequate except for the lack of reporting of how missing data
that increased attention to people with fatigue due to Guillain- were dealt with for an intention-to-treat analysis. We therefore
Barré syndrome may improve symptoms of fatigue. judged the overall risk of bias to be low.
The two studies evaluating ascorbic acid for Charcot-Marie-Tooth Pareyson 2011 reported a block-randomised, double-blind,
(CMT) disease 1A were block-randomised, double-blind, placebo- placebo-controlled study with a low risk of bias overall.
controlled trials of different doses of ascorbic acid for the treatment Since only one trial of amantadine for the treatment of fatigue in
of Charcot-Marie-Tooth disease 1A (Micallef 2009; Pareyson Guillain-Barré syndrome (Garssen 2006) and two trials of ascorbic
2011). The risk of bias for Micallef 2009 was low, despite the fact acid for treatment of Charcot-Marie-Tooth disease 1A (Micallef
that there was no report of how missing data were dealt with in the 2009; Pareyson 2011) are included in the review, it is not possible
intention-to-treat analysis. All other aspects of risk of bias assess- to address the review objectives for drug or other interventions.
ment were adequate. The main findings were that whilst ascorbic The Garssen 2006 study was relatively small (74 participants) and
acid was well-tolerated and safe for adults with Charcot-Marie- the results actually suggest that there may be an effect, but the study
Tooth disease 1A over a period of 12 months, there were no signif- is under-powered. We downgraded the evidence from this study
icant differences in effect across all outcomes, including fatigue, to low because of its imprecision (Summary of findings for the
between placebo, 1 g or 3 g of ascorbic acid (see Summary of main comparison). Larger studies including those with different
findings 2). There was a low risk of bias in the remaining study doses and frequency of administration of amantadine are desirable
(Pareyson 2011), and the main findings were that ascorbic acid before a definite conclusion about the effect of amantadine to treat
supplementation at a dose of 1.5 g per day for 24 months was fatigue in peripheral neuropathy and in particular in Guillain-
well tolerated but there was no significant effect on neuropathy or Barré syndrome can be given.
secondary outcomes compared with placebo. The studies of ascorbic acid for Charcot-Marie-Tooth disease 1A
were larger, and evaluated the effect of three different doses of
ascorbic acid in Charcot-Marie-Tooth disease 1A (Micallef 2009;
Overall completeness and applicability of Pareyson 2011). The lack of reporting of how missing data were
evidence dealt with in Micallef 2009 means that it is difficult to assess

whether the risk of bias is likely to have favoured the control or ropathies with pain as the primary presenting feature, and with po-
ascorbic acid treatment groups in this study . Nevertheless, the liomyelitis. There is therefore a small possibility that misdiagnosis
low risk of bias associated with the final study (Pareyson 2011) of participants may have affected study inclusion, and that stud-
and the fact that it has similar findings to Micallef 2009 meant ies with mixed populations of participants that include both sen-
that we did not downgrade the evidence because of trial design or sory, motor and combined sensory motor neuropathies as well as
conduct, but did downgrade it because of the lack of reporting of painful entrapment neuropathies or poliomyelitis were excluded.
how missing data was dealt with in Micallef 2009. This means that This could be a potential source of selection bias in the review.
there is moderate-quality evidence suggesting that ascorbic acid
has no benefit in the treatment of fatigue in CMT (See Summary Agreements and disagreements with other
of findings 2). studies or reviews
The primary outcome of fatigue was evaluated using the FSS or
VAS for fatigue in the included studies. The FSS is based on clas- There are currently no other systematic reviews of interventions
sical test theory (CTT), where scales may include items not rel- for fatigue in peripheral neuropathy available, despite many treat-
evant for particular clinical populations, and scale sum scores as- ments for fatigue including: drugs, pacing and grading of phys-
sume equal weighting of items where this has not necessarily been ical activity, general or specific exercise, compensatory strategies
demonstrated. This potentially introduces bias into the assessment such as orthotics, relaxation, counselling, cognitive and educa-
of the underlying outcome of interest. However, evidence in the tional strategies, either alone or as part of a multidimensional pro-
current review was not downgraded due to directness based on gramme.
outcomes used, since the effect of indirectness is not currently
known. Nevertheless, clinicians and researchers are moving to- Systematic reviews of interventions for fatigue
wards linearly weighted scales based on modern scientific methods in other conditions
such as Rasch models or item response theory (IRT) for evalu-
ation of outcomes in clinical populations. The FSS has recently In the absence of evidence from reviews or high-quality RCTs in
been fully evaluated and modified using the Rasch unidimensional peripheral neuropathy it may be helpful to consider the effect of
measurement model (RUMM2020) and is recommended for use interventions for fatigue in other conditions, including chronic
in immune-mediated neuropathy (van Nes 2009). Its relevance fatigue syndrome (CFS), fibromyalgia, multiple sclerosis and can-
for other peripheral neuropathies remains to be evaluated. cer-related fatigue. The high incidence of anxiety and depression
in functional somatic syndromes such as fibromyalgia and CFS
(Henningsen 2003) and the specific pathophysiological features
of multiple sclerosis and cancer-related fatigue mean that recom-
Potential biases in the review process
mendations from systematic reviews should be considered with
Two review authors were also investigators in an included trial caution. In all cases we included only evidence from Cochrane
(Garssen 2006), but neither was involved in assessing the risk of systematic reviews.
bias in this study.
We contacted study authors where there were missing data, and Drugs for fatigue
these was provided by Garssen 2006 for one secondary outcome. The evidence from Cochrane systematic reviews of drugs in the
We also contacted authors of published abstracts with limited re- treatment of fatigue in other conditions was of varying quality.
porting of data for further information, but they either did not A systematic review of serotonin and noradrenaline reuptake in-
reply or declined to, stating their intention to publish a full report hibitors (SNRIs) for fibromyalgia indicated that they only pro-
of the study in the future as a reason. vide a small improvement in fatigue (1.7% improvement) over
This systematic review used an adequate search strategy to avoid placebo (Hauser 2013). In reviews of amantadine or carnitine for
missing any RCTs on treatments for fatigue in peripheral neu- fatigue in multiple sclerosis, the poor quality of the available RCTs
ropathy. There were no language restrictions in our searches, and meant that no conclusion regarding their efficacy could be drawn
whilst a limitation of the review is that we did not initially plan in (Pucci 2009; Tejani 2012). Finally, whilst the updated system-
our protocol to search clinical trials registers, this additional search atic review of drug therapy for the management of cancer-related
was subsequently conducted; this difference between the protocol fatigue provides evidence that psychostimulant trials may give a
and review is flagged up below. clinically meaningful improvement in cancer-related fatigue, new
The criteria for types of participants for including studies accepted safety data indicate that the drugs are associated with increased
the diagnosis of peripheral neuropathy offered by the authors, pro- adverse outcomes and they can no longer be recommended in the
vided that it stipulated the presence of clinical impairment char- treatment of cancer-related fatigue (Minton 2010).
acteristic of peripheral neuropathy. We considered only sensory,
motor or combined sensory and motor peripheral neuropathies,
and excluded studies of participants with focal entrapment neu- Non-drug interventions for fatigue

We reviewed the evidence from recent Cochrane systematic re-
views of non-drug interventions for fatigue in other conditions,
which suggests that exercise (Edmonds 2004) and cognitive be-
havioural therapy (CBT) (Price 2008) for fatigue in CFS are both
better than usual care, with the evidence for CBT being more AUTHORS’ CONCLUSIONS
extensive and robust than for exercise. In a review of exercise for
treating fibromyalgia syndrome, there is very limited evidence that Implications for practice
aerobic-only exercise is likely to have no effect on fatigue in in-
dividuals with fibromyalgia (Busch 2007). A systematic review of The evidence from randomised controlled trials of the efficacy
cognitive behavioural therapies for fibromyalgia concludes that of interventions for fatigue in peripheral neuropathy is limited.
there was only a 5.8% (95% CI 0.05 to 11.3) effect on fatigue One small study comparing amantadine with placebo for fatigue
(Bernardy 2013). A systematic review of exercise for management in people with Guillain-Barré syndrome was not sufficiently pow-
of cancer-related fatigue concluded that aerobic exercise is benefi- ered to determine the effects on fatigue in this population. There
cial (Cramp 2012). was moderate quality evidence that the effects of ascorbic acid on
fatigue in people with Charcot-Marie-Tooth disease 1A are small.
Information about adverse effects was limited, although both treat-
Evidence from individual studies of interventions for ments appeared to be well tolerated in these conditions.
fatigue in neuropathy There was no evidence available from randomised controlled tri-
One article describing an uncontrolled case series of people with als to evaluate the effect of other interventions for fatigue in ei-
Charcot-Marie-Tooth disease 1A treated with modafinil for fatigue ther Guillain-Barré syndrome, Charcot-Marie-Tooth disease 1A
reported that all four participants had less fatigue within at least 10 or other causes of peripheral neuropathy.
days of commencing modafinil treatment (Carter 2006). The case
study design means that this report has substantial risk of bias but Implications for research
may suggest that modafinil is a potentially useful drug for fatigue The dearth of evidence from high-quality randomised controlled
in people with Charcot-Marie-Tooth disease 1A, and merits well- trials on the effects of interventions for fatigue in peripheral neu-
designed trials. ropathy indicates that treatments for this condition warrant fur-
The wide variety of causes and consequences in peripheral neu- ther research. Trials in relevant patient populations should have
ropathy, combined with a relatively low prevalence of individual adequate sample size, appropriate randomised controlled designs
conditions means that controlled trials of non-drug interventions and clinically relevant standardised and responsive measures of fa-
have been limited. In addition, despite the substantial prevalence tigue as an outcome.
of fatigue in people with peripheral neuropathy (Merkies 1999),
subjective fatigue is not routinely evaluated in RCTs and other
study designs of such interventions. However, one cohort study
(Garssen 2004) and one case-control study (Graham 2007) of exer-
ACKNOWLEDGEMENTS
cise for people with Guillain-Barré syndrome and chronic inflam-
matory demyelinating polyradiculoneuropathy (CIDP) reported The staff of the Cochrane Neuromuscular Disease Group (CN-
significant reductions in fatigue after 12 weeks of either supervised MDG).
aerobic cycling (Garssen 2004) or unsupervised community-based
The Trials Search Co-ordinator of the CNMDG developed search
exercise (Graham 2007). Whilst these studies are subject to very
strategies in collaboration with the review authors and provided
high risks of bias due to their design, they may suggest that RCTs
the literature searches.
of exercise for people with fatigue due to peripheral neuropathy
should be conducted. Editorial support from the CNMDG is supported by funding
from the MRC Centre for Neuromuscular Diseases.

REFERENCES
References to studies included in this review Attarian 2013 {published data only}
Attarian S, Dubourg O, Funalot B, Gonnaud P-M, Lacour
Garssen 2006 {published data only} A, Magy L, et al.A phase II randomized, placebo-controlled
Garssen MPJ, Schmitz PIM, Merkies ISJ, Jacobs BC, van multicenter clinical trial of three doses of PXT3003 in 80
der Meché FGA, van Doorn PA. Amantadine for treatment adult patients with CMT1A treated for 1 year. Journal of
of fatigue in Guillain-Barré syndrome: a randomised, the Peripheral Nervous System 2013;18(s2):S7–S8.
double blind, placebo controlled, crossover trial. Journal of
NCT00541164 {published data only}
Neurology, Neurosurgery and Psychiatry 2006;77(1):61–5.
NCT00541164. Effects of coenzyme Q10 on Charcot-
[PUBMED: 16361594]
Marie-Tooth disease. //clinicaltrials.gov/ct2/show/
Micallef 2009 {published data only} NCT00541164 (accessed 12 June 2014).
Micallef J, Attarian S, Dubourg O, Gonnaud P-M, Hogrel
J-Y, Stojkovic T, et al.Effect of ascorbic acid in patients NCT02121678 {published data only}
with Charcot-Marie-Tooth disease type 1A: a multicentre, NCT02121678. Effect of resistance and aerobic exercise
randomised, double-blind, placebo-controlled trial. Lancet in CIDP or MMN. www.clinicaltrials.gov/ct2/show/
Neurology 2009;8(12):1103–10. [PUBMED: 19818690] NCT02121678 (accessed 12 June 2014).
Pareyson 2011 {published data only} Ramdharry 2011 {published data only}
Pareyson D, Reilly MM, Schenone A, Fabrizi GM, Cavallaro Ramdharry GM, Pollard AJ, Anderson CA, Laurá M,
T, Santoro L, et al.Ascorbic acid in Charcot-Marie-Tooth Murphy SM, Hutton EJ, et al.Strengthening hip flexors to
disease type 1A (CMT-TRIAAL and CMT-TRAUK): a improve walking distance in people with Charcot-Marie-
double-blind randomised trial. Lancet Neurology 2011;10 Tooth disease. Journal of the Peripheral Nervous System.
(4):320–8. [PUBMED: 21393063] 2011; Vol. 16 (Suppl 3):S115–6.
References to studies excluded from this review Additional references
Burns 2009 {published data only} Bernardy 2013

Burns J, Ouvrier RA, Yiu EM, Joseph PD, Kornberg AJ, Bernardy K, Klose P, Busch AJ, Choy EHS, Häuser W.
Fahey MC, et al.Ascorbic acid for Charcot-Marie-Tooth Cognitive behavioural therapies for fibromyalgia. Cochrane
disease type 1A in children: a randomised, double-blind, Database of Systematic Reviews 2013, Issue 9. [DOI:
placebo-controlled safety and efficacy trial. Lancet Neurology 10.1002/14651858.CD009796.pub2]
2009;8(6):537–44.
Bigland-Ritchie 1978
Carter 2006 {published data only} Bigland-Ritchie B, Jones DA, Hosking GP, Edwards RH.
Carter GT, Han JJ, Mayadev A, Weiss MD. Modafinil Central and peripheral fatigue in sustained maximum
reduces fatigue in Charcot-Marie-Tooth disease type 1A: a voluntary contractions of human quadriceps muscle.
case series. American Journal of Hospice and Palliative Care Clinical Science and Molecular Medicine 1978;54(6):
2006;23(5):412–6. 609–14.
Enderlin 2008 {published data only} Bleijenberg 2003
Enderlin CA. Peripheral neuropathy, sleep and fatigue Bleijenberg G. Chronic fatigue and chronic fatigue
related to cancer and chemotherapy: preliminary findings. syndrome in the general population. Health and Quality of
Southern Online Journal of Nursing Research 2008;8(2):2p. Life Outcomes 2003;1:52.
NCT00484510 {published and unpublished data} Busch 2007
NCT00484510. A randomized, placebo-controlled, double Busch AJ, Barber KA, Overend TJ, Peloso PMJ, Schachter
masked 120 subject “futility design” clinical trial of ascorbic CL. Exercise for treating fibromyalgia syndrome.. Cochrane
acid treatment of Charcot Marie Tooth disease type 1A. Database of Systematic Reviews 2007, Issue 4. [DOI:
clinicaltrials.gov/ct2/show/NCT00484510 (accessed 21 10.1002/14651858.CD003786.pub2]
January 2013).
Cappelen-Smith 2000
Verhamme 2009 {published data only} Cappelen-Smith C, Kuwabara S, Lin CS, Mogyoros
Verhamme C, de Haan RJ, Vermeulen M, Baas F, de Visser I, Burke D. Activity-dependent hyperpolarization and
M, van Schaik IN. Oral high dose ascorbic acid treatment conduction block in chronic inflammatory demyelinating
for one year in young CMT1A patients: a randomised, polyneuropathy. Annals of Neurology 2000;48(6):826–32.
double-blind, placebo-controlled phase 11 trial. BMC
Medicine 2009;7(1):70–79. Chalder 1993
Chalder T, Berelowitz G, Pawlikowska T, Watts L, Wessely
References to ongoing studies S, Wright D, et al.Development of a fatigue scale. Journal of
Psychosomatic Research 1993;37(2):147–53.
Cohen 1989 Hauser 2013
Cohen RA, Fisher M. Amantadine treatment of fatigue Häuser W, Urrútia G, Tort S, Üçeyler N, Walitt B.
associated with multiple sclerosis. Archives of Neurology Serotonin and noradrenaline reuptake inhibitors
1989;46(6):676–80. (SNRIs) for fibromyalgia syndrome. Cochrane Database
of Systematic Reviews 2013, Issue 1. [DOI: 10.1002/
Cramp 2012
Cramp F, Byron-Daniel J. Exercise for the management 14651858.CD010292]
of cancer-related fatigue in adults. Cochrane Database Henningsen 2003
of Systematic Reviews 2012, Issue 11. [DOI: 10.1002/ Henningsen P, Zimmermann T, Sattel H. Medically
14651858.CD006145.pub3] unexplained physical symptoms, anxiety, and depression: a
meta-analytic review. Psychosomatic Medicine 2003;65(4):
DerSimonian 1986
528–33.
DerSimonian R, Laird N. Meta-analysis in clinical trials.
Controlled Clinical Trials 1986;7:177–88. Higgins 2003
Higgins JPT, Thompson SG, Deeks JJ, Altman DG.
Edmonds 2004
Measuring inconsistency in meta-analyses. BMJ 2003;327
Edmonds M, McGuire H, Price JR. Exercise therapy
(7414):557–60.
for chronic fatigue syndrome. Cochrane Database of
Systematic Reviews 2004, Issue 3. [DOI: 10.1002/ Higgins 2011a
14651858.CD003200.pub2] Higgins JPT, Green S (editors). Cochrane Handbook
for Systematic Reviews of Interventions Version 5.1.0
Elbourne 2002 [updated March 2011]. The Cochrane Collaboration,
Elbourne DR, Altman DG, Higgins JPT, Curtin F, 2011. Available from www.cochrane-handbook.org.
Worthington HV, Vail A. Meta-analyses involving cross-
over trials: methodological issues.. International Journal of Higgins 2011b
Epidemiology 2002;31(1):140–9. Higgins JPT, Deeks JJ (editors). Chapter 7: Selecting
studies and collecting data. In: Higgins JPT, Green S
Fisk 1994 (editors). Cochrane Handbook for Systematic Reviews
Fisk JD, Ritvo PG, Ross L, Haase DA, Marrie TJ, Schlech of Interventions Version 5.1.0 (updated March 2011).
WF. Measuring the functional impact of fatigue: initial The Cochrane Collaboration, 2011. Available from
validation of the fatigue impact scale. Clinical Infectious www.cochrane-handbook.org.
Diseases 1994;18(Suppl 1):S79–83.
Kaji 2000
Fisk 2002 Kaji R, Bostock H, Kohara N, Murase N, Kimura J,
Fisk JD, Doble SE. Construction and validation of a fatigue Shibasaki H. Activity-dependent conduction block in
impact scale for daily administration (D-FIS). Quality of multifocal motor neuropathy. Brain 2000;123(Pt 8):
Life Research 2002;11(3):263–72. 1602–11.
Gandevia 2001 Karlsen 1999
Gandevia SC. Spinal and supraspinal factors in human Karlsen K, Larsen JP, Tandberg E, Jorgensen K. Fatigue in
muscle fatigue. Physiological Reviews 2001;81(4):1725–89. patients with Parkinson’s disease. Movement Disorders 1999;
Garssen 2004 14(2):237–41.
Garssen MP, Bussmann JB, Schmitz PI, Zandbergen A, Kent-Braun 1993
Welter TG, Merkies IS, et al.Physical training and fatigue, Kent-Braun JA, Sharma KR, Weiner MW, Massie B, Miller
fitness, and quality of life in Guillain-Barré syndrome and RG. Central basis of muscle fatigue in chronic fatigue
CIDP. Neurology 2004;63(12):2393–5. syndrome. Neurology 1993;43(1):125–31.
Garssen 2007 Kleyweg 1991
Garssen MP, Schillings ML, van Doorn PA, van Engelen Kleyweg RP, van der Meché FGA, Schmitz PIM.
BG, Zwarts MJ. Contribution of central and peripheral Interobserver agreement in the assessment of muscle
factors to residual fatigue in Guillain-Barré syndrome. strength and functional abilities in Guillain-Barre syndrome.
Muscle & Nerve 2007;36(1):93–9. Muscle & Nerve 1991;14(11):1103–9.
GRADEpro 2008 Krupp 1989
Jan Brozek, Andrew Oxman, Holger Schünemann. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The
GRADEpro. 3.2 for Windows. Jan Brozek, Andrew Fatigue Severity Scale. Application to patients with multiple
Oxman, Holger Schünemann, 2008. sclerosis and systemic lupus erythematosus. Archives of
Neurology 1989;46(10):1121–3.
Graham 2007
Graham RC, Hughes RAC, White CM. A prospective study Krupp 2003
of physiotherapist prescribed community based exercise in Krupp LB. Fatigue in multiple sclerosis: definition,
inflammatory peripheral neuropathy. Journal of Neurology pathophysiology and treatment. CNS Drugs 2003;17(4):
2007;254(2):228–35. 225–34.
Lee 1991 Rosenberg 1988
Lee KA, Hicks G, Nino-Murcia G. Validity and reliability Rosenberg GA, Appenzeller O. Amantadine, fatigue, and
of a scale to assess fatigue. Psychiatry Research 1991;36(3): multiple sclerosis. Archives of Neurology 1988;45(10):
291–8. 1104–6.
Lennon 1993 Sanders 1996
Lennon SM, Koblar S, Hughes RAC, Goeller J, Riser AC. Sanders DB, Stalberg EV, Nandedkar SD. Analysis of the
Reasons for persistent disability in Guillain-Barré syndrome. electromyographic interference pattern. Journal of Clinical
Clinical Rehabilitation 1993;7(1):1–8. Neurophysiology 1996;13(5):385–400.
Martinez-Figuerora 1977 Schillings 2003
Martinez-Figueroa A, Hansen S, Ballantyne JP. A Schillings ML, Hoefsloot W, Stegeman DF, Zwarts MJ.
quantitative electrophysiological study of acute idiopathic Relative contributions of central and peripheral factors to
polyneuritis. Journal of Neurology, Neurosurgery and fatigue during a maximal sustained effort. European Journal
Psychiatry 1977;40(2):156–61. of Applied Physiology 2003;90(5-6):562–8.
Martyn 1998 Schünemann 2011a
Martyn C, Hughes RAC. Peripheral neuropathies. The Schünemann HJ, Oxman AD, Higgins JPT, Vist GE,
Epidemiology of Neurological Disorders. 1st Edition. London: Glasziou P, Guyatt GH. Chapter 11: Presenting results
BMJ, 1998. and ‘Summary of findings’ tables. In: Higgins JPT, Green
S (editors), Cochrane Handbook for Systematic Reviews
Merkies 1999
of Interventions Version 5.1.0 (updated March 2011).
Merkies IS, Schmitz PI, Samijn JP, van der Meché FG, van
The Cochrane Collaboration, 2011. Available from
Doorn PA. Fatigue in immune-mediated polyneuropathies.
www.cochrane-handbook.org.
European Inflammatory Neuropathy Cause and Treatment
(INCAT) Group. Neurology 1999;53(8):1648–54. Schünemann 2011b
Schünemann HJ, Oxman AD, Vist GE, Higgins JPT,
Minton 2010
Deeks JJ, Glasziou P, et al.Chapter 12: Interpreting results
Minton O, Richardson A, Sharpe M, Hotopf M, Stone P.
and drawing conclusions.. In: Higgins JPT, Green S
Drug therapy for the management of cancer-related fatigue.
(editors), Cochrane Handbook for Systematic Reviews
Cochrane Database of Systematic Reviews 2010, Issue 7.
of Interventions Version 5.1.0 (updated March 2011).
[DOI: 10.1002/14651858.CD006704.pub3]
The Cochrane Collaboration, 2011. Available from
Moss-Morris 2005 www.cochrane-handbook.org.
Moss-Morris R, Sharon C, Tobin R, Baldi JC. A randomized
Tamura 2007
controlled graded exercise trial for chronic fatigue syndrome:
Tamura N, Kuwabara S, Misawa S, Kanai K, Nakata M,
outcomes and mechanisms of change. Journal of Health
Sawai S, et al.Time course of axonal regeneration in acute
Psychology 2005;10(2):245–9.
motor axonal neuropathy. Muscle & Nerve 2007;35(6):
Pfeiffer 2001 793–5.
Pfeiffer G, Wicklein EM, Ratusinski T, Schmitt L, Kunze Tejani 2012
K. Disability and quality of life in Charcot-Marie-Tooth Tejani AM, Wasdell M, Spiwak R, Rowell G, Nathwani
disease type 1. Journal of Neurology, Neurosurgery and S. Carnitine for fatigue in multiple sclerosis. Cochrane
Psychiatry 2001;70(4):548–50. Database of Systematic Reviews 2012, Issue 5. [DOI:
Piper 1998 10.1002/14651858.CD007280.pub3]
Piper BF, Dibble SL, Dodd MJ, Weiss MC, Slaughter RE, van Nes 2009
Paul SM. The revised Piper Fatigue Scale: psychometric van Nes SI, Vanhoutte EK, Faber CG, Garssen M, van
evaluation in women with breast cancer. Oncology Nursing Doorn PA, Merkies IS, PeriNomS Study Group. Improving
Forum 1998;25(4):677–84. fatigue assessment in immune-mediated neuropathies: the
Price 2008 modified Rasch-built Fatigue Severity Scale. Journal of the
Price JR, Mitchell E, Tidy E, Hunot V. Cognitive behaviour Peripheral Nervous System 2009;14(4):268–78.
therapy for chronic fatigue syndrome in adults. Cochrane
References to other published versions of this review
Database of Systematic Reviews 2008, Issue 3. [DOI:
10.1002/14651858.CD001027.pub2] White 2009
Pucci 2009 White CM, van Doorn PA, Garssen MPJ, Stockley RC.
Pucci E, Brañas P, D’Amico R, Giuliani G, Solari A, Taus Interventions for fatigue in peripheral neuropathy. Cochrane
C. Amantadine for fatigue in multiple sclerosis. Cochrane Database of Systematic Reviews 2009, Issue 4. [DOI:
Database of Systematic Reviews 2009, Issue 1. [DOI: 10.1002/14651858.CD008146]
∗
10.1002/14651858.CD002818.pub2] Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Garssen 2006
Methods Randomised, double-blind, placebo-controlled, cross-over trial

Duration 18 weeks:
• 2 weeks pre-treatment period
• 6 weeks first intervention period
• 2 weeks washout
• 6 weeks second intervention period
• 2 weeks washout
Block randomisation in blocks of 6
Participants Identified from Dutch Guillain-Barré syndrome databank or the Dutch Neuromuscular
Disorders Patients Association (Vereniging Spierziekten Nederland)
80 (40 female and 40 male) neurologically stable participants with GBS (no apparent
change in GBS disability score within 3 months of the start of the study) with severe
fatigue (mean FSS score of ≥ 5) who were over 18 years old, with a GBS disability score
< 3 (able to walk at least 10 metres with or without aid)
Exclusion criteria: presence of severe fatigue prior to onset of GBS, comorbidity with
conditions that might cause fatigue, receiving medication that may induce fatigue, score
of > 10 on the depression subscale of the HADS, contraindications for treatment with
amantadine including pregnancy and breastfeeding, renal dysfunction, and allergy to
amantadine
Interventions Amantadine 100 mg once daily for 1 week then twice daily for 5 weeks (n = 40 allocated,
36 in analysis) or placebo (n = 40 allocated, 38 in analysis), followed after washout by
the opposite intervention
Outcomes 1. Reduction in severe fatigue by at least 1 point on the FSS

2. Impact of fatigue - FIS
3. Mood - HADS
4. Participation/handicap - RHS
5. Quality of life - SF-36 and EQHQ
Funding This study was supported by a grant from the Dutch Organization for Scientific Research,
NWO (grant no. 940-38-009)
Declarations of interest None stated
Notes 7 people were excluded prior to randomisation for insufficient fatigue (< 5 on FSS)
Ethical approval was gained in May 2000 but no information on the duration of the
study is reported
Risk of bias
Bias Authors’ judgement Support for judgement

Garssen 2006 (Continued)
Random sequence generation (selection Low risk Allocation sequence generated by block
bias) randomisation (block size of 6)
Allocation concealment (selection bias) Low risk Allocation sequence list sent to pharmacy
department and consecutive people at out-
patient department assigned a number
from 1 to 80 in sequence of entering the
study
Blinding (performance bias and detection Low risk Placebo-controlled trial

bias)
All outcomes
Incomplete outcome data (attrition bias) Low risk Analysis based on 74 participants, as 6
All outcomes withdrawn: 3 from first intervention period
(1 from the placebo group and 2 from the
amantadine group) and a further 3 partic-
ipants withdrawn due to inadequate com-
pletion of FSS
Selective reporting (reporting bias) High risk Only P values reported for 4 secondary out-
comes: fatigue impact, mood, participation
limitation, and quality of life
Other bias Low risk Washout period similar to studies of aman-

tadine in other populations. Carry-over ef-
fect of 2-week washout period tested by au-
thors and no significant effect reported
Micallef 2009
Methods Block-randomised, double-blind, placebo-controlled, multicentre trial

Duration: 12 months
Participants Participants were recruited from 3 hospital sites in France

195 patients were initially screened and 179 (110 women and 69 men) with a clinical
diagnosis of CMT1A confirmed by genotyping with duplication in 17p11.2 were sub-
sequently randomised prior to intervention
Participants were aged between 18 and 70 years of age and had to have at least one
motor symptom or sign (gait disorder, distal amyotrophy, foot deformation or muscle
weakness). People with any cause of neuropathy other than CMT1A and other medical
conditions: kidney stones, hypersensitivity or intolerance to ascorbic acid or lactose, un-
corrected oxalosis, cardiovascular disease, poliomyelitis or radiotherapy, those for whom
ascorbic acid supplementation was contra-indicated, women of childbearing age who
were pregnant, breastfeeding or not using adequate contraception, were excluded

Micallef 2009 (Continued)
Interventions Participants were randomised in a 1:1:1 ratio to ascorbic acid 1 g, (n = 56), ascorbic acid 3
g (n = 61) or placebo containing lactose (n = 62), all given as a single daily administration
of 3 hard gelatine capsules for 12 months
Outcomes Primary outcome: Charcot-Marie-Tooth neuropathy score (CMTNS) at 12 months after

commencement of intervention
Secondary outcomes
1. Muscle strength - QMT
2. Gait velocity - timed 10 metre walk
3. Activity - ODSS
4. Quality of life - Medical Outcomes SF-36
5. VAS for fatigue, pain, cramps and difficulty walking
6. CGI-S
Funding No indication of funding given
Declarations of interest One study author (MF) is one of the inventors of the patent for ascorbic acid for the
treatment of CMT1A licensed to Murigenetics. Two of the study authors (MF and OB)
are scientific advisers to Murigenetics and are 15% shareholders in the company
Notes Participants were recruited from Sept 2005 to September 2007
Risk of bias
Random sequence generation (selection Low risk Block randomisation stratified to site and
bias) sex for subgroup analyses with sequence
generation using an independent contract
research organisation
Allocation concealment (selection bias) Low risk Randomisation list sent to an indepen-
dent research organisation for preparation
of ascorbic acid and placebo capsule (iden-
tical hard gelatine capsules)
Blinding (performance bias and detection Low risk Ascorbic acid and placebo formulated from
bias) hard gelatine capsules (identical in weight,
All outcomes appearance and taste). Placebo contained
lactose
Incomplete outcome data (attrition bias) Low risk Analyses were on an intention-to-treat ba-
All outcomes sis. Reporting of dropouts (total 16) was
effective: 6 withdrew (1 in placebo, 4 in
1g ascorbic acid and 1 in 3 g ascorbic acid
group) due to adverse events (cystitis, myal-
gia, gastralgia, depression, colitis and in-
somnia), 2 were lost to follow up (1 in

Micallef 2009 (Continued)
placebo and 1 in 3 g ascorbic acid group)

, 6 withdrew consent (1 in placebo, 4 in 1
g ascorbic acid and 1 in 3 g ascorbic acid)
, 1 was non-compliant with treatment in
the placebo group and 1 placebo partici-
pant was missing a baseline CMTNS score
Selective reporting (reporting bias) Low risk All stated primary and secondary outcomes
were fully reported
Other bias Unclear risk Intention-to-treat analysis meant sufficient

sample size for adequate power. No indica-
tion of how missing data were dealt with for
intention-to-treat analyses. Therefore diffi-
cult to determine risk of bias due to method
Pareyson 2011
Methods Parallel, randomised, double-blind, placebo-controlled, multicentre trial

Duration 24 months
Participants Recruited from 8 Italian hospitals and 1 UK hospital between March 2006 and Sept
2007
354 patients screened for eligibility and 277 eligible patients subsequently randomly
assigned to intervention. Six participants did not subsequently receive the allocated in-
tervention and were excluded, leaving intervention 138 (78 women and 60 men) and
placebo 133 (85 women and 48 men). All participants had a clinical and genetic diagno-
sis of CMT1A. Participants were aged between 18 and 70 years and had a CMTNS of
between 1 (excluding the electrophysiology component) and 35 (including the electro-
physiology component). Exclusion criteria: any cause of neuropathy other than CMT1A;
other clinically significant neurological or systemic disease; limb surgery in the 6 months
before screening or planned before the final assessment; pregnancy, breastfeeding or
planning to become pregnant during the study; receipt of extra-dietary ascorbic acid
supplementation in the 3 months prior to screening; any contraindications to ascorbic
acid (nephrolithiasis, glucose-6-phosphate dehydrogenase deficiency or iron overload)
Interventions Ascorbic acid at a dose of 1.5 g per day as 3 x 0.5 g tablets (1 at breakfast and 2 at dinner)
(n = 138) or placebo tablets of identical appearance, taste and smell (n = 133)
Outcomes Primary outcome: CMTNS at 12 months after commencement of intervention

Secondary outcomes:
Muscle strength - MVIC using hand-held dynamometer for: hand-grip, 3-point pinch,
ankle dorsiflexion and plantarflexion
Pain - VAS
Fatigue - VAS
Activity - ONLS, 10 metre walk test, 9-hole peg test
Quality of life - Medical Outcomes SF-36

Pareyson 2011 (Continued)
Funding This trial was supported by Telethon-UILDM (grant numbers GUP04002 and
GUP05007) and AIFA (Grant number FARM53APAH) in Italy, Muscular Dystrophy
Campaign (MDC RA3/736/1), CMTUK, UKMRC and NIHR BRC in the UK
Bracco SpA (Milan) provided ascorbic acid and placebo tablets
Declarations of interest One of the study authors (RACH) was a consultant for Octopharma, Baxter, Laboratoires
Francais de Fractionnement et des Biotechnologies (LFB) and Novartis. One study author
(ASo) was a board member of Novartis and received speaker’s honoraria from Sanofi-
Aventis
Notes
Risk of bias
Random sequence generation (selection Low risk The randomisation sequence was com-
bias) puter-generated (pseudo-random number
generator) by an independent randomisa-
tion unit
Allocation concealment (selection bias) Low risk Randomisation sequence known only to
independent unit and the drug dispen-
sary for preparation of identically packaged
ascorbic acid and placebo tablets. Treat-
ment was allocated centrally by telephone
and stratified by centre and disease severity
(CMTNS ≤ 10 versus > 10 or CMTES ≤ 8
versus > 8) with a block size of 4 unknown
to the investigators in each centre
Blinding (performance bias and detection Low risk Participants, treating physicians and physi-
bias) cians assessing outcomes were blinded to
All outcomes treatment allocation throughout study
Incomplete outcome data (attrition bias) Low risk Missing data for 22 participants who were
All outcomes lost to follow-up were imputed according
to Rubin’s multiple imputation approach.
Reasons for loss to follow-up were reported
in full with similar numbers in each group
for each reason: 13 participants had adverse
events (7 in ascorbic acid; 6 placebo), 6
withdrew their consent (3 in each group)
, 2 in the ascorbic acid group moved away
from the study area and 1 in the placebo
group left for personal reasons
A further 8 participants stopped treatment
but did attend follow-up so no imputed

Pareyson 2011 (Continued)
data were used for these participants
Selective reporting (reporting bias) Low risk All stated primary and secondary outcomes
fully reported at 24 months
Other bias Low risk None
Incomplete outcome data (attrition bias). Summary assessment of risk of bias based on risk of bias associated with unclear description
of how missing data was managed in intention-to-treat analysis
CGI-S: Clinical Global Impression Severity Scale
CMT: Charcot-Marie-Tooth
CMTES: Charcot-Marie-Tooth Examination Score
CMTNS: Charcot-Marie-Tooth Neuropathy Score
EQHQ: Euroquol Health Questionnaire
FIS: Fatigue Impact Scale
FSS: Fatigue Severity Scale
GBS: Guillain-Barré syndrome
HADS: Hospital Anxiety and Depression Scale
MVIC: Maximal Voluntary Isometric Contractions
ODSS: Overall Disability Sum Score
ONLS: Overall Neuropathy Limitations Scale
RHS: Rotterdam Handicap Scale
Short Form 36 Health Survey
VAS: visual analogue scale
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Burns 2009 RCT of ascorbic acid for CMT1A in children. Fatigue not used as an outcome
Carter 2006 Not a RCT - 4 single case series of fatigue reduction associated with modafinil treatment
Enderlin 2008 Not a RCT - observational study of sleep disturbance in multiple myeloma
NCT00484510 No relevant outcomes to be evaluated (only CMT neuropathy score and mRNA). Fatigue not used as an outcome
Verhamme 2009 RCT of ascorbic acid treatment for CMT1A in people under 25. Fatigue not used as an outcome
CMT: Charcot-Marie-Tooth; mRNA: messenger RNA; RCT: randomised controlled trial

Characteristics of ongoing studies [ordered by study ID]
Attarian 2013
Trial name or title A phase II randomized, placebo-controlled multicenter clinical trial of three doses of PXT3003 in 80 adult
patients with CMT1A treated for 1 year
Methods Randomised double-blind placebo-controlled trial of 12 month intervention
Participants 80 participants between 18 and 65 years with DNA-proven CMT1A
Interventions 4-arm trial:

• 3 intervention arms of different dosages (low 1/100, intermediate 1/50, high 1/10) delivered as 5 ml
liquid doses twice daily of PXT3003 (a new low-dose combination of 3 already approved drugs (naclofen,
naltrexone and D-sorbitol)). In the highest dose, the active ingredients were 1/0 and 1/100 of the currently
used registered indication dosages.
• 5 ml liquid doses twice daily of placebo
Outcomes Primary objective was clinical and biological safety. Secondary objectives were to obtain preliminary data on
clinical and functional outcomes (including CMTNS, ONLS, 6MWT, 9HPT, VAS for fatigue and pain,
QMT - hand grip and ankle dorsiflexion and clinical global impression) and electrophysiological parameters
(including sensory and motor responses of median and ulnar nerves - NCV, DL and amplitude)
Starting date December 2010
Contact information S Attarian, Hopital La Timone
Notes Primary objective of determining clinical and biological safety achieved. Analyses of secondary outcomes
ongoing
NCT00541164
Trial name or title Effects of coenzyme Q10 on Charcot-Marie-Tooth disease
Methods Randomised, double-blind, cross-over trial with open-label follow-up study
Participants 18 to 75 years of age with CMT confirmed by medical record review. Regarding weakness, fatigue, and pain,
subjects must report experiencing maximum levels of 3.0 cm or more on the 10 cm visual analog scale (VAS)
for any 2 of the 3 symptoms over the past month
Interventions Coenzyme Q10, 600 mg/day (300 mg twice/day for 24 weeks) or chewable placebo twice/day for 24 weeks
followed by the alternative treatment in the cross-over period
Outcomes Primary: changes in weakness, fatigue and pain (60 weeks)

Secondary: improvements in quality of life (60 weeks); blood serum levels of the oxidised and reduced forms
of CoQ10 (60 weeks); liver function tests (visits 1, 6 and 12)
Starting date September 2007

NCT00541164 (Continued)
Contact information Sharon Plank, MD. John P Murtha Neuroscience and Pain Institute, Johnstown, Pennsylvania, United States,
15904
Notes Study completion date January 2013
NCT02121678
Trial name or title Effects of coenzyme Q10 on Charcot-Marie-Tooth disease
Methods Randomised, parallel assignment, open-label
Participants Aged 18 to 80 years with definite or probable CIDP or definite or probable MMN of over 6 months’ duration,
diagnosed according to EFNS/PNS criteria. On stable maintenance dose of intravenous immunoglobulin
Interventions Experimental: run-in - aerobic - resistance

• Week -12 to 0: no training
• Week 0 to 12: aerobic training (ergometer bicycle, twice weekly, increasing to 3 times weekly.
Participants to train at 60 - 75 % of maximal oxygen consumption (VO2-max) for 30 minutes per training
session)
• Week 12 to 24: resistance training (dumbbells: week 0 to 4: 70% to 80% of 1-repetition max (1-RM);
week 5 to 8: 75% to 86 % of 1-RM; week 9 to 12: 80% to 92% of 1-RM)
Experimental: run-in - resistance - aerobic
• Week -12 to 0: No training
• Week 0 to 12: Resistance training (dumbbells as above)
• Week 12 to 24: Aerobic training (ergometer bicycle as above)
For resistance training, participants train unilaterally, the opposite site serves as reference
• MMN participants train elbow and wrist (flexion/extension)
• CIDP participants train knee and elbow (flexion/extension)
Outcomes Primary (measured at -12, 0, 6, 12 weeks or -12, 0, 12, 18, 24 weeks):

• changes in isokinetic muscle strength
• changes in VO2-max
Secondary (measured at -12, 0, 6, 12 weeks or -12, 0, 12, 18, 24 weeks). Change in the following measures:
• Medical Research Council (MRC)
• change in ODSS
• change in 6-Minute Walk test
• change in Quality of Life measured by The Short Form (36) Health Survey (SF-36) questionnaire
• change in FSS
Starting date April 2014
Contact information Department of Neurology, Aarhus University Hospital. Contact: Henning Andersen, DMSc
Department of Neurology, Rigshospitalet, Copenhagen University Hospital: Contact: Ingelise Christiansen
Notes

Ramdharry 2011
Trial name or title Strengthening hip flexors to improve walking distance in people with CMT disease
Methods Single-blinded, cross-over study. Duration of intervention 16 weeks
Participants 32 participants with CMT disease. Mean age 43.5 (SD 14.76) years with median CMTES of 10
Interventions 16-week home-based programme of training to increase hip flexor muscle strength
Outcomes Hip flexor muscle strength measured using fixed myometry at 4 angles (0, 20, 45 and 90 degrees). Secondary
outcomes of walking endurance, gait speed, exertion, fatigue and general activity not reported in abstract
Starting date Not reported
Contact information Ramdharry GM, MRC Centre for Neuromuscular Diseases, Department of Molecular Neurosciences, UCL
Institute of Neurology, London
Notes Study completed, awaiting authors’ reporting of outcomes in full publication prior to next update of systematic
review
6MWT: 6 metre walk test

9HPT: 9-hole peg test
CIDP: chronic inflammatory demyelinating neuropathy
CMT: Charcot-Marie-Tooth
DL: distal latency
EFNS/PNS: European Federation of Neurological Sciences/Peripheral Nerve Society
FSS: fatigue severity score
MMN: multifocal motor neuropathy
NCV: nerve conduction velocity
ODSS: Overall disability sum score
ONLS: Overall Neuropathy Limitations Scale
QMT: quantitative muscle testing
PMP22 mRNA: peripheral myelin protein 22 (PMP22) messenger RNA
PXT3003: a low-dose combination of naclofen, naltrexone and D-sorbitol
VAS: visual analogue scale

DATA AND ANALYSES
Comparison 1. Ascorbic acid versus placebo
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Change in fatigue > 12 weeks 2 404 Std. Mean Difference (IV, Fixed, 95% CI) -0.12 [-0.32, 0.08]
intervention (measured at 12
to 24 months) (VAS 0 - 10 or 0
- 100)
2 Change in 10 m timed walk > 2 406 Mean Difference (IV, Fixed, 95% CI) -0.39 [-1.04, 0.26]
12 weeks after commencement
of the intervention (measured
at 12 to 24 months)
3 Change in speed of completion 1 Mean Difference (IV, Fixed, 95% CI) Subtotals only
of 9 hole peg test > 12 weeks
months)
4 Change in physical function 2 400 Std. Mean Difference (IV, Fixed, 95% CI) 0.08 [-0.12, 0.28]
score of SF-36 > 12 weeks
after commencement of the
to 24 months)
5 Change in mental function 1 Mean Difference (IV, Fixed, 95% CI) Subtotals only
score of SF-36 > 12 weeks
after commencement of the
months)
6 Serious adverse events 2 450 Risk Ratio (M-H, Fixed, 95% CI) 0.70 [0.39, 1.27]
ADDITIONAL TABLES
Table 1. Data and analysis: amantadine versus placebo
Outcome Studies Participants Statistical method Effect estimate
Change in participation 1 (cross-over) 74 Mean difference (IV, Fixed, 95% 0.31 [-0.09 to 0.72]
(using the Rotterdam CI)
Handicap Scale) > 12
weeks after commence-
ment of intervention

WHAT’S NEW
Last assessed as up-to-date: 5 November 2013.
Date Event Description
18 December 2014 Amended Correction to plain language title
CONTRIBUTIONS OF AUTHORS
CMW, MG and RCS screened lists of abstracts and titles from electronic and manual searches, CMW and RCS extracted data
from included study, CMW prepared first draft of the review and the other authors (PAvD, MPJG and RCS) reviewed the draft.
CMW incorporated comments after discussion, where appropriate and the final draft was returned to all authors for approval prior to
submission.
DECLARATIONS OF INTEREST
One author (CW) is currently conducting a single-blind randomised controlled trial of tailored home exercise for people with inflam-
matory neuropathy (Home exercise for Inflammatory Neuropathy Trial - HINT). The trial is funded by the Guillain-Barré syndrome
support group UK (GBSSG). CW is on the Medical Advisory Board of the GBSSG UK. She has no known conflicts of interest.
Pieter A van Doorn (PAvD) and his institution have received consultancy fees from Talecris and CSL Behring for membership of the
Scientific board of the ICE trial in CIDP and a scientific board on IVIg in chronic polyneuropathy. PAvD’s department has received
research grants from Baxter, Sanquin and Talecris: 1. A grant to conduct a RCT comparing IVIg versus IVIg and steroids in Guillain-
Barré syndrome; 2. A grant to conduct a RCT investigating the effect of a second course of IVIg (SID-trial) in Guillain-Barré syndrome
patients with a poor prognosis; 3. A grant to conduct a prospective international study on the effect of a second course of IVIg in
Guillain-Barré syndrome patients with a poor prognosis. MPG and PAvD have co-ordinated or conducted one of the trials that was
eligible for inclusion in the review (Garssen 2006).
Marcel P Garssen has no known financial conflicts of interest.
Rachel Stockley has no known conflicts of interest.
SOURCES OF SUPPORT
Internal sources
• Kings College London, UK.

External sources
• No sources of support supplied
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Our protocol stated that we would consider trials for inclusion where interventions for fatigue formed part of a multidimensional
programme; however, we removed this inclusion criterion.
We removed PEDRO from the list of resources searched.
Use of funnel plots was not appropriate given the small number of studies included.
We revised the potential subgroup analyses, removing age. We also added a definition of ’significant weakness’.
We revised the section Dealing with missing data, which originally stated ’If participant drop out leads to missing data, we shall
conduct an intention-to-treat analysis. We shall contact trial authors or sponsors of studies to provide missing statistics such as standard
deviations. For dichotomous outcomes, we shall regard participants with missing outcome data as treatment failures and include these
in the analysis. For continuous outcomes, we shall carry forward the last recorded value for participants with missing outcome data.’
The section was revised since two of the included studies Micallef 2009; Pareyson 2011 undertook an intention-to-treat analysis where
missing data were reported. The third study (Garssen 2006) was a cross-over trial that reported reasons for withdrawing participants
with missing data effectively, and where data were analysed for the review as the number of participants responding favourably to the
intervention versus placebo in either period; imputation of data was not appropriate,
We did an additional search of clinical trials registries, for ongoing trials.
We included more detail in our adverse events outcome, subdividing it into “any adverse events, adverse events which lead to discontin-
uation of treatment and serious adverse events, which are those which are fatal, life-threatening, or require prolonged hospitalisation.”
INDEX TERMS
Medical Subject Headings (MeSH)

Amantadine [adverse effects; ∗ therapeutic use]; Ascorbic Acid [adverse effects; ∗ therapeutic use]; Charcot-Marie-Tooth Disease
[∗ complications]; Fatigue [etiology; ∗ therapy]; Guillain-Barre Syndrome [∗ complications]; Peripheral Nervous System Diseases [com-
plications]; Randomized Controlled Trials as Topic
MeSH check words

Adult; Child; Humans


Interventions For Fatigue in Peripheral Neuropathy

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Interventions For Fatigue in Peripheral Neuropathy

Uploaded by

Copyright:

Available Formats

Interventions for fatigue in peripheral neuropathy (Review)

White CM, van Doorn PA, Garssen MPJ, Stockley RC

Interventions for fatigue in peripheral neuropathy (Review)

Interventions for fatigue in peripheral neuropathy (Review) i

Interventions for fatigue in peripheral neuropathy

Claire M White1 , Pieter A van Doorn2 , Marcel PJ Garssen3 , Rachel C Stockley4

Editorial group: Cochrane Neuromuscular Disease Group.

Data collection and analysis

PLAIN LANGUAGE SUMMARY

Interventions for fatigue in peripheral neuropathy (Review) 3

Amantadine versus placebo for fatigue in Guillain-Barré syndrome

Patient or population: people with fatigue in Guillain-Barré syndrome

Assumed risk Corresponding risk

Reduction in fatigue of 257 per 1000 162 per 1000 OR 0.56 74 ⊕⊕ -

Change in activity limi- - - - - - -

Health related quality of - - - - - EHQ and SF-36 measured

GRADE Working Group grades of evidence

Interventions for fatigue in peripheral neuropathy (Review) 6

Types of participants Search methods for identification of studies

Interventions for fatigue in peripheral neuropathy (Review) 7

Interventions for fatigue in peripheral neuropathy (Review) 8

Interventions for fatigue in peripheral neuropathy (Review) 9

Interventions for fatigue in peripheral neuropathy (Review) 10

Risk of bias in included studies

Interventions for fatigue in peripheral neuropathy (Review) 11

Interventions for fatigue in peripheral neuropathy (Review) 12

Fatigue after more than 12 weeks

Interventions for fatigue in peripheral neuropathy (Review) 13

Participation restrictions after 12 or more weeks

Health-related quality of life (HRQoL) after 12 or more

Interventions for fatigue in peripheral neuropathy (Review) 14

Interventions for fatigue in peripheral neuropathy (Review) 15

Ascorbic acid versus placebo for fatigue in peripheral neuropathy

Patient or population: people with fatigue in peripheral neuropathy

Assumed risk Corresponding risk

Placebo Ascorbic acid

Change in fatigue 12 or - The mean change in fa- - 404 ⊕⊕⊕ -

Overall Neuropathy Lim-

Health-related quality of - The mean health-related - 400 ⊕⊕⊕ -

GRADE Working Group grades of evidence

Interventions for fatigue in peripheral neuropathy (Review) 19

Interventions for fatigue in peripheral neuropathy (Review) 20

Interventions for fatigue in peripheral neuropathy (Review) 21

References to studies excluded from this review Additional references

Burns 2009 {published data only} Bernardy 2013

Interventions for fatigue in peripheral neuropathy (Review) 24

Characteristics of included studies [ordered by study ID]

Methods Randomised, double-blind, placebo-controlled, cross-over trial

Outcomes 1. Reduction in severe fatigue by at least 1 point on the FSS

Declarations of interest None stated

Bias Authors’ judgement Support for judgement

Interventions for fatigue in peripheral neuropathy (Review) 25

Blinding (performance bias and detection Low risk Placebo-controlled trial

Other bias Low risk Washout period similar to studies of aman-

Methods Block-randomised, double-blind, placebo-controlled, multicentre trial

Participants Participants were recruited from 3 hospital sites in France

Interventions for fatigue in peripheral neuropathy (Review) 26

Outcomes Primary outcome: Charcot-Marie-Tooth neuropathy score (CMTNS) at 12 months after

Funding No indication of funding given

Notes Participants were recruited from Sept 2005 to September 2007

Bias Authors’ judgement Support for judgement

Interventions for fatigue in peripheral neuropathy (Review) 27

placebo and 1 in 3 g ascorbic acid group)